Mostly color perception

First question: What it reality? That's an entire subject area of philosophy called "epistemology". The Nature paper does not try to address that question.

Einstein is the one who invoked the word "realism" in his original 1935 EPR paper, but he was using it in a very special sense viz., to shut the door on the noted quantum mechanicians of the day: mostly Bohr, Heisenberg and Pauli. He wanted to convince them (and anyone else) that forming quantum pairs in a randomly correlated ("mixed") state could not *realistically* permit their quantum numbers to remain correlated after physical separation. Quantum mechanics (QM) allows that they can, but Einstein wanted to show (by Kantian synthetic a priori reasoning) that this outcome was intolerable because it was "unrealistic" (especially on the grounds of causation) to have "action at a distance" between physically separated randomized quantum pairs such that, if I measure one of the pair after separation, then the quantum numbers of the other will still be found to be correlated. Even by today's standards, this is an extremely powerful argument. For the quantum mechanicians in 1935, the EPR paper came like "a bolt out of the blue" and Pauli insisted that Bohr respond quickly in case confidence in QM started to erode.

Aside: Einstein was more than justified in his EPR argument because of his earlier triumph: gravity. Newton's law of gravitation is precisely an *action-at-a-distance* model. Exactly how is the gravitational force that holds the earth in its orbit around the sun propagated through space? (Let's not get into the earth, moon, sun trio) When Newton was asked about this force in comparison to his other "impulse" force (F = ma), he stated "Hypothesis non fingo" or "I don't have to," in today's vernacular. If you were Newton you could get away with that kind of thing. So, Einstein answered for him. Gravity is a field quantity, like Maxwell's classical E's and B's, that corresponds to warping in space-time itself. In other words, gravity was a classical field, and it eliminated the notion of action-at-a-distance.

Vindicated in this way, Einstein believed that QM could be explained ultimately with a classical, causal theory, if we were just smart enough to figure it out. The first and foremost of such attempts was the so-called hidden-variables (HV) theory due to Bohm in the 1950's. Einstein later declared Bohm's original local, hidden-variables (HV) theory to be a "cheap shot"; this from the man who most desired that such things be true, in order to maintain his view of (classical) reality! But, by that time, Einstein also wanted any QM explanations to include gravity and electromagnetism; somewhat along the lines of today's quantum field theories. From that vantage point, a modified, non-relativistic Schrödinger equation does look a little cheap.

Experimental testing of Bell's Theorem (especially since the 1980's; that's 50 years after Einstein's EPR argument), has essentially excluded all local HV theories; which Bell incorrectly expected to be proven true. Here, "local" means that the wavefunction is uniquebe ly defined at each point in space-time. If the EPR pair correlations could explained classically, then standard probability theory would predict certain numbers. All EPR-like experiments show that the classical predictions are off by exactly a quantum factor of root-2. Bell even thought that Einstein's EPR argument was far more rigorous than anything Bohr ever uttered. Be that as it may, nature doesn't care who's doing the arguing. So, Einstein was wrong, but for the right reasons. I think he deserves a lot of credit for that.

My rather cursory reading of the Zeilinger et al., paper indicates that some people (including Tony Leggett; who is no slouch, let alone a crackpot) have examined non-local HV models for any Einsteinian wiggle room. Their experiments show that even these more subtle HV variants (or a large class of them) are also excluded as alternative explanations of QM. In other words, Bohr, Heisenberg and Pauli are as correct as they ever were, and in particular, light is a purely quantum phenomenon; not classical waves or classical particles. This, I believe, was Prof. Beretta's point (Abandon all hope, ye of a classical mindset). We know the correct quantum rules, we may not like them but they work to better than 1 part in a billion accuracy. That's the most accurate theory ever devised by mankind, but it does not match the classical physics by which our brain and eyes (and thus our comprehension of reality) have evolved.

Posted by RocketRoo on 5/21/2007 3:48 PM
» Permalink 

See http://www.sciam.com/article.cfm?articleID=37F1485E-E7F2-99DF-3FD5D372EB2E6D43&sc=I100322

Because of stability issues with entangled states, biphoton transmission usually takes place in a lab or down an optical fiber. Zeilinger has another goal of transmitting and receiving biphotons between a ground station and the International Space station (about 200 miles). The military must be lapping this stuff up.

Posted by RocketRoo on 5/21/2007 9:52 PM
» Permalink 

As for forgetting all you were taught about the physics of refraction and wave theory (whether Huygens or Helmholtz), I would say that is not necessary, but it's fun to try. What do I mean by that? It turns out that, not only do we need the photon to understand all the super-duper quantum optical effects like lasing and entanglement, but we can also understand the everyday optical effects like refraction and diffraction in terms of the photon. The photon provides the correct description and solutions for all these situations. Normally, the everyday optics is not presented in terms of the photon because geometric rays or scalar waves are sufficiently good approximations. This is analogous to the fact that we still use Newtonian mechanics for solving everyday engineering problems (like repairing melted freeways) even though we know it's fundamentally "wrong". It breaks down for strong gravitational fields and atomic-scale fluctuations.

Similarly, if you need to include polarization effects, then you need the vector fields of Maxwell's equations, which btw are the correct relativistic equations of motion for the photon. That's because the photon is a piece of quantum mechanics that just happened to fall into the middle of the 19th century (but most people haven't noticed). If you would like to see how the photon description of everyday optical effects works, I would highly recommend reading Chapter 2 of Feynman's Mautner lectures http://www.amazon.com/QED-Strange-Theory-Light-Matter/dp/0691024170/ref=pd_bbs_sr_1/102-6198625-1544965?ie=UTF8&s=books&qid=1179805168&sr=1-1 As far as I know, he's the only physicist who has tried to do this. He even explains why Newton got a lot (most) of his opticks wrong.

If I may, I would like to take up your comment of 5/20/2007. I must say I didn't quite get it. It seems to be an indirect reference to coherence length. Perhaps you could clarify further, as time permits. In the meantime, I'm inclined to think the statement "...where light ceases to behave like a photon and becomes ... waves..." looks like an interesting question but is actually a false dichotomy. We (our eyes) are not "aware" of either particle-like or wave-like qualities in light. It's just a stimulus that causes us to 'see'. Maybe I missed your point, but permit me to ramble on a little further.

The concept of photons confers a "graininess" to light. If I take the position that all light is comprised of photons (as I did above), I could ask more legitimately about the controlled conditions under which I might detect (though not directly with my eyes) the graininess of light. We know from single-photon interference measurements (as well as with electrons) that it takes about 10-50,000 events for enough 'grains' to accumulate (say, in a storage scope) so as to produce recognizable interference 'fringes'. In other words, what we see is the time-averaged effect of many photons falling on the retina. A 100 Watt light globe, for example, puts out about Avogadro's number of photons per second, so we are always seeing in a time-averaged fashion. This accounts for what Technician Beretta is saying at the end of his comment of 5/21/2007 9:54 PM.

This effect is largely independent of coherence length. The visibility of the fringes is maximized with a monochromatic, fully, phase-coherent primary source. As the primary becomes less coherent ("partially coherent" per Born & Wolf), the relative phases of the photons become more randomized and the fringe visibility falls, because there are multiple fringe patterns (one for each frequency) superimposed upon each other. In daylight (or any thermal primary), the fringes become completely smeared out because the relative phases are completely randomized in that case.

Posted by RocketRoo on 5/21/2007 11:19 PM
» Permalink 

Yes, the eye is a rather poor optical device and the photons have to traverse the whole retina to reach the rods and cones, but we are averaging over a very large number of photons. Photons of a certain energy will make it to the receptors at a different rate than photons at a different energy. However, when we consider the receptor sensitivities, we measure at the cones, i.e., the rest of the eye is ignored.

The effect of a photon depends on its energy, which is Plank's constant time its frequency. To tune the photon absorbtion according to its energy, the visual protein molecule has attached to its backbone a chromophore through a lysine. In humans there are four different chromophores, which are the pigments that give an energy dependent sensitivity. They are rhodopsin (rods), cyanolabe (S-cones), chlorolabe (M-cones), and erytholabe (L-cones). The catch probability distributions are not delta functions, but broad distributions, so there is an overlap and lightness can be perceived in addition to hue.

The opsin molecules are not one per cone or rod. Rather, each cone or rod is a stack of membranes, each with many opsines. The neurotransmitter signal flows outside the photoreceptor, thereby summing the catches over all membranes and in each membrane summing all individual photon catches.

When we are interested in the total spectral sensitivity of the eye, then we cannot use physiology because we do not yet understand it well enough. Instead, we do psychophysical experiments. The resulting absorption probabilities are the color matching functions.

As for low power coherent light, coherence does not affect a photon's energy, which depends only on the frequency. Therefore, the perception is not different. Of course, we can build instruments that depend on coherence, so we can see different views, but this does not change how the eye itself operates.

I suppose if I write Pi = (e+e-) for positronium [has to be capital pi, since lower case 'pi' is a meson = (quark-antiquark) pair], then what they have seen is Pi_2. Their formal paper will appear in Sept. 13 issue of Nature.

This is interesting for another reason having to do with entanglement and coherence; the subjects of this blog thread.

Positronium is basically unstable, and when it decays by falling into itself (like falling down a set of quantum stairs) it usually gives off 1,2,3, ... photons (depending on the number of stairs). The most common decay channel is 2 photons. John Wheeler (he of the so-called "delayed-choice" interferometer, amongst other things) suggested in c.1945 that these photons should have complimetary polarizations. In fact, they were the first entangled photons produced in the lab c.1949 by Wu and Shaknov at Columbia Univ. In today's lingo, they are type-II entangled.

Because of the annihilation energy involved, however, these are gamma-ray photons. So, we have the odd situation where it is "easier" to produce entangled gamma-photons than coherent gamma-photons! That's where the Pi_2 comes in. The diatomic form occurs on a silica (sand) substrate. One goal is to get enough of these groupings on the substrate to form a BEC (see http://h20325.www2.hp.com/blogs/color/archive/2007/06/24/3691.html). That, it seems, would allow one to have more than one source emitting simultaneously and therefore phase-coherently. Voila! The gamma-ray laser.

From this I can't tell how what the binding orbitals are, how the diatoms bind to the substrate or what temperatures apply. Perhaps someone who takes a look at the Nature paper when it comes out, can report on that.

Posted by PublicPassport on 9/12/2007 6:15 PM
» Permalink