Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

How can sending a massive chunk of carbon through the double-slit prove

  1. Jan 8, 2010 #1
    ...The superposition of two states?...
    When I send a massive, classical chunk of carbonous stuff through a double slit, you can't expect me to believe that that chunk will magically turn into two equally massive chunks and do its constructive/destructive interfering, yet this is exactly what I'm being told by certain of current physicists; this is their explination for the diffraction pattern being seen.
    The massive chunk of classical matter in this case, of course, being the buckyball.
     
  2. jcsd
  3. Jan 8, 2010 #2

    ZapperZ

    User Avatar
    Staff Emeritus
    Science Advisor
    Education Advisor
    2016 Award

    The "buckyball" in question isn't JUST a "classical matter". Look at under what condition it has been put under to get all parts of it to be in coherence with each other. This is a property of a quantum particle! Look at the Delft/Stony Brook experiment that had to deal with an entity consisting of 10^11 particles being in a superposition of states!

    This question is also puzzling because I have no idea if you have a problem with the whole concept of superposition in QM, or if you simply have a problem with accepting only the interpretation of THIS particular experiment. If it is the latter, then you need to re-read the actual paper again and see the parameters of the experiment.

    Zz.
     
  4. Jan 8, 2010 #3
    Thanks for the answer.
    Is this a matter of coherence of the buckyball, or one of the role of coupling of the noise-free system (buckyball, in this case)?
    Is it possible then to create a noise-free, coherent cat and perform Schrodinger's experiment for real--using a bona-fide feline?
    You also say that certain of conditions had been set up to get all 10^11 particles of the buckyball to be in coherence, then call it a "property of a quantum particle." Woudn't it be correct to say, "quantum particles?"

    My primary issue is the role of superposition in the slit experiment, exclusively.
     
  5. Jan 8, 2010 #4

    DrChinese

    User Avatar
    Science Advisor
    Gold Member

    Are you really wanting to split these hairs? ZapperZ knows of what he speaks...

    At any rate, it isn't unusal usage to see a system in a specific non-factorable state referred to as a quantum particle instead of "particles" if the individuals are not distinguishable.
     
  6. Jan 8, 2010 #5
    I'm not denying ZapperZ knows what he's talking about (which would explain my continuing discourse with him).
    One thing about fullerine diffraction is that the buckyball is heated to a very hi temp before being sent thru the first slit. How come?
     
  7. Jan 8, 2010 #6

    ZapperZ

    User Avatar
    Staff Emeritus
    Science Advisor
    Education Advisor
    2016 Award

    No, it is NOT correct to say that, because all of those 10^11 particles make up one large "macro particle" that are in a superposition of states. In that case, it is a macro particle that are having two simultaneous current states.

    You really should read the original papers, not just for that Delft/Stony Brook experiment, but for that buckyball experiment as well.

    Zz.
     
  8. Jan 8, 2010 #7
    It'd be great if you guys would direct me to an excellent link regarding buckyball diffraction and its superposition implication specifically. That would, of course, be appreciated...
    (BTW, I tried to find reading related to the Delft/Stony Brook experiment and related matter and could only find this:
    With something as large as C60 you'd think we can see both superposed balls simultaneously.
     
    Last edited: Jan 8, 2010
  9. Jan 8, 2010 #8
    Last edited by a moderator: Apr 24, 2017
  10. Jan 8, 2010 #9

    f95toli

    User Avatar
    Science Advisor
    Gold Member

    The work by Stony Brook and Delft that ZappersZ is refering it quite old (I don't even think Stony Brook are working on superconducting qubits anymore). However, there are LOT of groups working on this now and many new papers are published every month so there are no shortage of papers to read.

    This review is also pretty old (2005) but it does give an overview of the field of superconducting qubits (how they are entangled etc)

    http://arxiv.org/abs/cond-mat/0508729
     
  11. Jan 8, 2010 #10
    The 10^11 figure is extremely misleading. In reality there are only a few hundred single particle states that have different occupation numbers in the two parts of the superposition. What is going on here is that the Cooper pairs consist of electrons on opposite sides of the Fermi sphere. A current running one way can be pictured as shifting the whole Fermi sphere in one direction. You can then have 10^11 Cooper pars participating in the supercurrent. If the current runs in the other direction, the Fermi sphere is shifted in the opposite direction and then you can say that you have again 10^11 Cooper pairs in that supercurrent.

    However, the overlap between the two Fermi-spheres is almost 100%; almost all the occupied single particle states in the current running in one direction are also occupied when the current is running in the other direction.
     
  12. Jan 8, 2010 #11
    The superposition state allows us to calculate probabilities. In the double slit experiment, if we determine which slit the quantum particle passed through, i.e. we measure the position of the quantum particle at the slits, then the superposition state gives the probability of passing through each slit. If, on the other hand, we measure the momentum of the quantum particle at the slits, the superposition state gives the probability that the particle will be scattered at a given angle. etc. The probability you get depends on the observable being measured.

    But, the important thing is that the superposition state allows us to calculate ONLY probabilities. No experiment has ever shown a single particle passing through both slits, or a half of the particle passing through each slit. The superposition state does not tell us what the particle is doing, as classical physics does.
     
  13. Jan 8, 2010 #12

    f95toli

    User Avatar
    Science Advisor
    Gold Member

    But the most remarkable thing with solid state qubits is that they are so big (tens of microns) and you can easily see them in an ordinary micrscope, in circuit where two or more qubits are coupled via a resonator the circuit is several mm in size. Hence, it is really an example of macroscopic quantum effects.

    Also, a slightly technical detail, in phase and flux qubits it is the phase that is the variable, and the phase is a macroscopic property of all the Cooper pairs in the conductor.
     
  14. Jan 9, 2010 #13

    ZapperZ

    User Avatar
    Staff Emeritus
    Science Advisor
    Education Advisor
    2016 Award

    No, your interpretation of it is wrong. Read Leggett's treatment of it[1].

    When you form a supercurrent, the WHOLE GLOB is one coherent entity because it has condensed to only one state! That is why many QM experiments are done in the superconducting state - the quantum effects can be seen at the macro scale. The formation of the coherence gap in the SQUID experiment is a result of a dynamics superposition of the supercurrent - the whole glob having current in opposite direction simultaneously, not simply a fraction of it.

    Zz.

    [1] A.J. Leggett, J. Phys. Condens. Matt., v.14, p.415 (2002).
     
  15. Jan 9, 2010 #14
  16. Jan 19, 2010 #15
    You're other reply was very smart, but must we result to "rigor" when trying to prove a point? In a way, that's like the articulate philosopher--when questioned about a questionable, abstract, and nebulous subject-matter--becomes even more verbose!
    Verbose to the point of deliberately confusing his audience.
     
    Last edited: Jan 19, 2010
  17. Jan 19, 2010 #16
    Now here's my language! That amorphous and invisible line of scrimmage between the quantum and the classical. When will we dissolve this useless and anti-pragmatic line! I say, that as Physicists, we must do with all speed; current research of the S/S qbits and fullerine diffraction is now demanding it of us.

    This means a necessary absolution of the concept of superposition. Superposition is strictly statistical in nature, I know, but classical mechanics is resolutely anti-statistics.

    We must consider abandoning certain of aspects of statistics in our science even if Einstein prefered it sometimes.
     
  18. Jan 19, 2010 #17
    Now I'm kind of understanding what you're saying here.
    The usual QM answer to the classical question, "What happens when we fire the single photon thru the two slits? Huh? The next photon will seem to interfere with it. How? It was fired thru seconds later!"
    The QM answer is this: "Well, the single photon created a double of itself (superposed) and interfered with itself at the slit. That's why we see the diffraction fringes. There was true, almost classical interference at the slits, since we did have two [albeit superposed] photons there."

    Now this is my extent of current thought (don't blame me; blame popular media since that's where I derived my info).

    Your suggestion is interesting, snce you invoke the HUP into the matter; something I haven't read about. So is it a matter of looking at momentum of photon v. position of photon statistically? And if this is so, and since we know the HUP to be something of a fundamental 'property' (sorry), then how to truly account for the diffraction fringes? Is there a set of equations (non-rigorous, of course), that you can direct my attention span?
     
  19. Jan 20, 2010 #18
    The difficulty, as I see it, is that we refuse to accept the quantum observations as the final say in the matter. We want to know “how the experiment works”. We keep demanding a classical description for a quantum event - and there isn’t any!

    In the mechanical universe of classical physics every event has a cause that is known to us. If something happens we can explain “why it happened.”

    Not so in the quantum domain! Quantum mechanics does not describe processes or the behavior of particles moving through space-time. It only tells us the possible results of a measurement and the statistical distribution of those results. Unfortunately, most of us refuse to accept such meager scraps of information.

    Quantum events have no classical explanation. That is why we invented quantum mechanics. But we still cling to classical concepts. For example, we still believe the classical maxim, “Waves exhibit interference, particles do not.” So, if particles exhibit interference, as they do in some quantum events, then they must be acting like waves. Then the weirdness begins, as you have noticed.

    You want to know how quantum interference happens when particles pass through two slits one at a time. We know there is no classical explanation and quantum mechanics doesn’t tell us either. So, there is no way to answer your question! The truth is, we have no idea “what is really happening” when “a massive chunk of carbon passes through the slits”!

    There is a quantum explanation however. At the slits, we know the particle’s position, which introduces an uncertainty in its momentum. Repeating the experiment, subsequent particles can then be scattered at different angles, i.e. with different momenta. Repeating the experiment many times we get a probability distribution of scattered particles that has maxima and minima reminiscent of constructive and destructive interference in wave optics.

    Further, no one has ever observed that, “--------the single photon created a double of itself (superposed) and interfered with itself at the slit.” Your understanding of the superposition state is erroneous. In the double slit experiment we have a superposition of position eigenvectors, not a superposition of particles.

    There is a good discussion in arxiv.org/pdf/quant-ph/0703126. before the calculations begin.
    Best wishes.
     
  20. Jan 20, 2010 #19
    Thanks for the reply. I always knew I was wrong about the, "How can one photon make a copy of itself and then interfere with itself?"

    But, its true we know position of the photon at the slit but not its momentum (conversely, if we know the photon's momentum, we'll knever know when it reaches a slit, and which one at that). This is the HUP in action.
    But still, how can this eventual statistical distribution of multiple photon firings consecutively produce a diffraction pattern? Why not a simple gaussian/Bessel function distribution on the screen, as classicalists were expecting out of the Stern-Gerlach experiment?
    I can't see how knowledge of the photon position at the slit with cluelessness about the photon's momentum at the slit can ultimately lead to a nice diffraction pattern in the end!
    E.Q.'s (9) thru (14) in this link: arxiv.org/pdf/quant-ph/0703126 can't be of any help (read concluding remarks between eq.'s (12) and (13))

    Also, fullerine diffraction uses boulders of carbon that are so huge as they pass thru the slit, that we should be able to get something like a 10M/pixel camera at Target for like $199.00 on sale to take a picture of it all. I mean, the buckyball is so huge; huge enough to make its position at the slit a non-issue I'd think, and a well-resolved camera pic of that buckyball going thru the slit. As for determining the buckyball's momentum before the pic is taken can't be much harder than finding out the momentum of dropping a grain of sand onto a scale or something...

    The CEO of Target can take credit for introducing groundbreaking physics equipment at their stores for a $199.00 sale price.

    Concluding remarks in the article can't really be convincing by stating the authors didn't result to wave optics on the unshown superposition of the delta functions (discuss!), or by referring to the Born Postulate. Remember, eq's (9)-(13) implied an infintesmally-thin slit, which is absurd. This impractical approach to quantum interference seems to discredit matters.

    We must discuss the superpsed delta functions, i'd think.

    Thoughts, anyone?
     
    Last edited: Jan 20, 2010
  21. Jan 21, 2010 #20

    ZapperZ

    User Avatar
    Staff Emeritus
    Science Advisor
    Education Advisor
    2016 Award

    The HUP gives you, in some sense, only a qualitative description of the single-slit diffraction. It tells you why there is such a spread in the perpendicular momentum. It cannot tell you why there is such a pattern, because it wasn't "built" with such information. To get that, you have to go back to ground zero (the same starting point that gives the HUP as a consequence) with the Hamiltonian. Check the http://arxiv.org/ftp/quant-ph/papers/0703/0703126.pdf" [Broken] that I've cited numerous times to see how you derive single, double, and multiple slits pattern using QM without invoking classical wave optics.

    In fact, one can even extend this approach into 2D as well without invoking classical optics. See Frank Rioux treatment of http://www.users.csbsju.edu/~frioux/diffraction/ej33n1.pdf" [Broken].

    Zz.
     
    Last edited by a moderator: May 4, 2017
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook




Similar Discussions: How can sending a massive chunk of carbon through the double-slit prove
Loading...