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

Atom Interferometry for Nuclear Reactions

  1. Jul 13, 2007 #1
    Recently, we've heard about research into storing anti-hydrogen as a BEC. Atom lasers are touted for their potential to make ultra-precise measurements, including at the quantum scale, because of their small DeBroglie wavelengths in comparison to light.

    Once, I emailed Dr Wolfgang Ketterle immediately after he was awarded the Nobel Prize, and asked him about whether Bose-Einstein Condensate could be used to somehow achieve nuclear fusion. He good-naturedly replied back, chuckling that no, it was not possible, because the density of the BEC is far too low.

    Alright, a single BEC/superatom might not be able to do this, but what about 2 or more BECs in relative motion to each other -- ie. 2 atom lasers made to interfere with each other?

    My point in speculating about this idea, is that so far in our attempts at achieving nuclear fusion, we have been trying to get a bunch of particles to collide with each other. But what if you could manipulate matter as a wave, and make two such waves very sharply interfere with each other?

    What if you had 2 very "powerful" atom lasers -- in this case, "power" would probably refer to mass throughput or KE -- which would be configured to maximize collisions between their respective atoms resulting in nuclear reactions.

    If fusion reactions wouldn't produce enough power, then could atom lasers be used for conveniently producing matter-antimatter collisions?

    Last edited: Jul 13, 2007
  2. jcsd
  3. Jul 13, 2007 #2


    User Avatar
    Science Advisor


    Quantum mechanics doesn't work like that. You can't "get around" or "get over" the
    Coulomb barrier by some type of "interference".

    In essence, the waves of a particle can only interfere with other waves from the SAME
    particle - not other particles. Remember the magnitude of the wave function gives the
    probability density function for finding the particle at a given point in space in the limit
    of the Born approximation.

    You don't get a modification of the probability because another particle's wave function
    happens to overlap. There has to be some type of interaction potential for that to
    happen. It's like two radio waves of different frequencies that criss-cross in space.

    Just because at some point in space, one radio wave is "+" and the other wave is "-";
    doesn't mean that the two waves annihilate each other. Nope - they pass right
    through each other.

    You do consider the quantum mechanical wave nature of the problem when you
    compute the reaction cross-section; because one considers the "tunneling" of the
    Coulomb barrier.

    Dr. Gregory Greenman
  4. Jul 13, 2007 #3
    Hi Greg, that wasn't quite what I was saying.

    Clearly you can overcome Coulombic repulsion using momentum. The purpose of using the atom lasers is to have colliding streams of atoms (each of these atoms having momentum), but where the collision of the atoms isn't a hit-and-miss affair, and is instead more coherently organized.

    The momentum or kinetic energy is what breaches the Coulomb barrier, the interference or synchronization is what keeps you on target. What's the point of 2 oncoming atoms having lots of momentum or KE, if they miss? Every miss is a wasted opportunity.

    That's why I said atom laser, because atom laser means moving atoms, but moving in a coherent manner, just like coherent light moves in a coherent manner. Lasers can be synchronized.
  5. Jul 13, 2007 #4
  6. Jul 14, 2007 #5


    User Avatar
    Science Advisor


    I don't think you are going to be able to control the trajectory of the
    individual ions to the point where you can dictate a collision.

    No - just make and confine a plasma of the requisite density and temperature;
    and you will have all the collisions you need.

    For example, in a nuclear fission reactor, we don't control whether or
    not an individual neutron hits a nucleus or not. The collisions are all
    stochastic. Besides, the number of collisions number in the thousands
    of billions - you aren't going to be able to control that many particles.

    But what's more - you don't have to.

    Dr. Gregory Greenman
  7. Jul 15, 2007 #6
    But Greg, all those wasted collisions are what make the reactors so hazardous and heavy. All those dangerous neutrons and the shielding/blanketing they necessitate.

    Photonic lasers of course all travel at C.
    But with atom lasers we can control how fast the atoms travel. So what is the limit on how fast the atoms in an atom laser can be made to travel?

    Research in photonic lasers is trying to make them more powerful.
    I presume the analogy for atom lasers would be to try and make them have more KE.
    High-KE atom lasers might be more useful in practical applications, because they would allow you to direct a large amount of KE in a precisely controlled way.
  8. Jul 15, 2007 #7


    User Avatar
    Staff Emeritus
    Science Advisor

    The collisions in nuclear processes are inherent in the nature of the process. What makes reactors so 'potentially' hazardous is the high levels of energy involved. The confinement systems - pressure vessels and piping of a fission reactor system or confinement magnets and vessel for a fusion reactor - require massive strutures.

    Laser systems are equally massive. Just look at the Shiva system, and now the National Ignition Facility.


    The term 'photonic laser' is redundant.

    Particle beams and lasers require a lot of energy input compared to the useful energy produced. Trying to use lasers or beams on an atomic level is inherently inefficient and counter productive. One CANNOT control nuclear collisions on a nuclear level, let alone the atomic level. Just look at the cross-sectional dimension of a laser beam!
  9. Jul 15, 2007 #8


    User Avatar
    Science Advisor


    BALONEY!!! What "wasted collisions" are you talking about - that's NONSENSE!!

    Reactors are NOT hazardous. The neutrons have NOTHING to do with
    the reactors being heavy. The fuel - Uranium - is the heaviest naturally
    occurring element. Additionally, the reactor has heavy shielding. However,
    the bulk of that is to shield gammas not neutrons.

    As Astronuc points out - the term "photonic lasers" is redundant. ALL lasers;
    which stands for Light Amplified by Stimulated Emission of Radiation are based
    on Photons; i.e. "light". Even their predecessor the MASER - for Microwaves
    Amplified by Stimulated Emission of Radiation are ALSO photons.

    You don't have "coherent laser beams" of particles - that's NOT a laser.

    Particle accelerators get nucleons up to 99.999..% of the speed of light;
    which is the ultimate speed limit.

    There's NO SUCH THING as an "atom laser". Lasers are based on the process
    called "stimulated emission" - that is one photon can trigger the release of
    another photon and the two photons will be in phase. There's no analogous
    process for atoms - "atom laser" is a NONSENSE term.

    There are already machines that do that - and they are NOT lasers.

    Cyclotrons, synchrotrons, and other particle accelerators are used in
    nuclear physics research. Plasma torches are used to cut metal, diamond...
    many materials very precisely.

    Dr. Gregory Greenman
  10. Jul 15, 2007 #9
    Cutting things is not the purpose that I was alluding to.

    I'm saying that BECs-in-motion (aka "atom lasers") would allow us to collimate and collide greater numbers of nuclei with greater frequency, and greater accuracy.

    Nuclei could be targetted at each other with less chance of missing, and therefore make more efficient use of inputted energy.

    http://www.npl.washington.edu/AV/altvw97.html (scroll to section on atom lasers and 4-wave mixing)

    Look, a low-mass particle, whether in motion or at rest, is going to have some jiggle to it -- the DeBroglie wavelength, a quantum effect due to the buffeting of the dynamic vacuum. So if you're trying to collide it with a similarly low-mass particle, then that jiggle is going to create some uncertainty on whether you get a collision or not.

    But with a BEC-superatom, you've got the mass of multiple atoms concentrated into a single superatom, and so it's going to have much less jiggle, since the dynamic vacuum can't buffet its heavier mass as easily. So that means you can aim that superatom more precisely when trying to achieve a collision with another oncoming superatom. So you're much more likely to achieve a collision.

    Chemical reactions have been achieved by colliding opposing streams of atom laser beams. Each input/collision stream consisted of a chemical reagent in BEC form, while a 3rd (output) stream was generated, made of the chemical reaction product.

    So then the question becomes one of trying to accelerate your superatoms to enough velocity to overcome Coulombic repulsion, so that collisions become meaningful for nuclear reaction purposes. It's a question of relative velocity for the colliding beams.

    How fast does a superatom have to be travelling relative to another superatom it's colliding with, in order to overcome the Coulombic barrier? Let's assume a superatom of Deuterium.
    Last edited: Jul 15, 2007
  11. Jul 15, 2007 #10
  12. Jul 15, 2007 #11


    User Avatar
    Staff Emeritus
    Science Advisor

    It is unfortunate that a professional would use the term 'atom laser', when it has nothing to do with lasers. One is talking about a monoenergetic and coherent beam of atoms colliding with some fixed target or opposing beam.

    One still cannot avoid scattering - it is inherent in the physics of particle interaction. One cannot control atoms in the keV range down the level of the radius of a nucleus, and the atom is several orders of magnitude larger.

    Using atoms involves a more complex interaction than using nuclei.
  13. Jul 15, 2007 #12
    When we try to collide atoms with each other for fusion purposes, we like to exploit the Coulombic shielding that neutrons cause, so that we can more easily obtain the collision. That's why we like D and T over H. Ideally, you'd like the neutrons to be positioned on the forward-facing sides of your respective nuclei as they rush to hit each other, so that they will provide the Coulombic shielding.

    But we're forced to rely upon statistical effects, because we can't control which way nuclei are oriented as they hit each other. Or can we?

    With BECs and atom lasers, our atoms are all in phase with each other. What works for one will work for all, because they're all synchronized. With atomic interferometry, we could play with the beams and adjust some parameter like beam path (just like Michelson-Morley did) until we get our atom beams colliding in just the right way. We'll know that we've calibrated our colliding beams optimally, once we observe optimal reaction products (ie. nuclear reaction products)

    I'm wondering if this might even be useful for muon-catalyzed fusion? After all, we want our muons to diatomically bind with hydrogen nuclei in the most efficient way, so that the short-lived muons will catalyze the most nuclear fusions possible before they die. BEC hydrogen is about the most dense hydrogen you can get, which might facilitate the muons visiting the most hydrogens possible. Just another thought.
  14. Jul 15, 2007 #13
    But according to the concept of "matterwave", the properties of the atom beam have cyclical periodicity. There is a wavelength.

    How can you control a laser beam of photons to etch pits on a DVD? The laser beam has the wave properties.
    How can you do laser interferometry? Lasers have those wave properties.

    In the case of an atom laser, the wavelength is far smaller than the wavelength of light, because of the higher mass of the atoms. Likewise, those atoms have kinetic energy.

    Exactly what is the relationship between kinetic velocity of the atoms, and their uncertainty/imprecision?
    Does acceleration automatically cause decoherence? What is the critical threshold of beam decoherence below which we cannot obtain sufficient numbers of nuclear collisions between our opposing atom laser beams?

    When you say 'scattering' -- in 4-way mixing of atom beams that scattering is a result of collision. Fine, we don't have to mind scattering of the reaction products, after our collisions have been achieved.
    Last edited: Jul 15, 2007
  15. Jul 15, 2007 #14
    Nextly, why couldn't you apply photonic lasers to further optimize collision conditions between the atoms of opposing atom laser beams?

    We know that femtosecond/attosecond-pulse lasers are capable of distorting the electron cloud of an atom. Doing so could have an effect on the orientation of the charged nucleus. What if you could synchronize/arrange your photonic lasers (femtosecond/attosecond-pulse) to interact with the atom laser beams and manipulate the orientation of their nuclei, in order to optimize their chances of breaching the Coulomb barrier?

    So as your colliding atoms approach each other, the femtosecond/attosecond pulses would arrive, distorting their electron clouds so that they bunch up between the nuclei to provide more shielding, thus allowing the colliding nuclei to successfully overcome the Coulomb barrier and fuse.

    That way, you're bringing multiple physics tricks into play at once, to get you to the goal. You're using photoelectric effect (cloud distortion), you're using wave nature of matter (mass-superpositioning for jiggle reduction and trajectory refinement), so that you can fight Coulomb's Law.
    Last edited: Jul 15, 2007
  16. Jul 15, 2007 #15
    http://www.nist.gov/public_affairs/releases/n99-09.htm [Broken]

    Okay, so making an atom laser beam more intense would mean having more nuclei travelling in that beam, which would hopefully increase the possibility of nuclear collisions. With more nuclei being in phase with each other along the beam trajectory, it could allow us to synchronize their collision trajectories to maximize the collisions between the opposing beams.
    Last edited by a moderator: May 3, 2017
  17. Jul 15, 2007 #16
    http://www.laserfocusworld.com/display_article/292386/12/ARCHI/none/News/ATOM-OPTICS:-Slow-helium-source-is-first-step-toward-atom-interferomete" [Broken]

    So you guys were talking about tokamak plasma having a low density, but if a fast-moving atom beam pulse has an internal temperature of 50mK, that sounds like it corresponds to a high density. Of course we'd want to have the most intense beam possible, to maximize the number of nuclei smacking into each other. The more crowded our collision zone is, the better.
    Last edited by a moderator: May 3, 2017
  18. Jul 15, 2007 #17

    Okay, so a few tens of nanometers is still quite a lot of dispersion relative to the radius of an atomic nucleus. But when you're talking about a whole lot of atoms travelling down that beamline, you might achieve quite a number of collisions within that cross-sectional area.

    The greater your beam velocity and the greater your beam intensity, the more your atomic throughput ("power"), and the more collisions you could expect.
    Last edited by a moderator: Apr 22, 2017
  19. Jul 16, 2007 #18


    User Avatar
    Science Advisor


    WRONG!!!! Neutrons DON'T sheild the Coulomb force.

    Your concept of having the neutronn on the "forward facing" side is also NONSENSE!!!

    You have to think of the neutrons in the nucleus as if they were in ORBITS - the
    "Tinker-Toy" model of the nucleus where the protons and neutrons are represented as
    discrete balls is WRONG!!!

    Neutrons don't shield the Coulomb force. What is happening is that the strong nuclear
    force is MUCH more powerful than the Coulomb force - but it is short range. The idea
    is to give the particles enough energy so that by the time the Coulomb force stops them;
    they are close enough so that the strong nuclear force can "overpower" the Coulomb
    force and fuse the two particles into a single nucleus. The strong nuclear force can do
    this after all - it's what is holding the nucleus together - a nucleus that may have many,
    many protons - all of which are repelling each other.

    Neutrons bring with them a certain amount of nuclear "binding energy". Protons also
    bring nuclear "binding energy" - but they also bring Coulombic "unbinding energy".

    So neutrons - because they increase the nuclear force without adding to the repulsion
    forces - can act like a nuclear "glue". Neutrons DON'T shield the Coulomb force - they
    provide more nuclear force to overcome the Coulomb force.

    Dr. Gregory Greenman
  20. Jul 16, 2007 #19


    User Avatar
    Science Advisor


    NOPE - your concept of how the quantum mechanics of colliding beams works is just
    plain WRONG!!!

    You CAN NOT DO what you are suggesting.

    The only reason you can do what Michaelson and Morley did is because you essentially
    have photons that are interfiering with THEMSELVES!!!

    If you take a Michaelson and Moreley interferometer, and increase the length of one
    of the legs so that the difference between the lengths of the two legs exceeds the
    "coherence length" of the photons - then you STOP getting interference effects because
    you are no longer having the same photons interfere with themselves.


    You are essentially applying principles of wave interference that are good for photons
    that are interfering with themselves; to a regime where you have discrete particles that
    do NOT share the same wavefunction.

    Dr. Gregory Greenman
    Last edited: Jul 16, 2007
  21. Jul 16, 2007 #20


    User Avatar
    Science Advisor


    Any given muon can only catalyze exactly ONE fusion reaction!!!

    This is NOT like a catalyst that can be used over and over again.

    Muons are a ONE SHOT catalyst - so this is all NONSENSE!!

    Dr. Gregory Greenman
  22. Jul 16, 2007 #21
    Here is a reply from Dr William Phillips, 1997 Nobel Prize Winner in Physics

    Dr Phillips then also further replied:

  23. Jul 16, 2007 #22
    I'm sure that this isn't true. IIRC, a single muon can catalyze an average of 103 nuclear D-D fusion reactions before it decays/dies. Let me find a reference:
    I'm wondering if muons could catalyze the fusion better if coupled to a BEC superatom?
  24. Jul 16, 2007 #23


    User Avatar
    Staff Emeritus
    Science Advisor

    No. One is not going to get 103 reactions before a muon decays - half-life ~ 2 microseconds or so. One might be fortunate to get a d and t or d traveling nearly parallel (highly improbable, but could happen) and the muon just happens to pass by at the right moment (even more improbable). One is lucky to get one reaction.

    One needs a better reference than a Wikipedia article.


    Dr. Phillips may be recalling accelerator driven transmutation - and that is not based on 'atomic lasers'.

    This is something entirely different. One should look at the size of those 'pits' on a DVD in comparison with an atom. And one is simply changing the optical properties (reflectivity) of the surface - it either reflects or does not. It's on/off - hence a digital (binary) system - and the laser is fixed in the vertical and azimuthal - sliding back and forth radially. This is also low power stuff.

    I think one is mixing apples, oranges and bananas here.
    Last edited: Jul 17, 2007
  25. Jul 16, 2007 #24
    Regarding decoherence of a laser, may I point out that the more massive atoms are less vulnerable to decoherence from quantum effects (ie. propagation through the vacuum) than the far lighter photons are.

    So I would see superatoms as having far greater decoherence length than photons.

    Atom lasers have been shown to have interference patterns -- these occur when they are interfering with themselves.

    An atom laser beam can be split, and then re-collided with itself.
  26. Jul 16, 2007 #25
    Muons have a lifespan of over 2 microseconds:


    "The muon (from the letter mu (μ)--used to represent it) is an elementary particle with negative electric charge and a spin of 1/2. It has a mean lifetime of 2.2μs, longer than any other unstable lepton, meson or baryon except for the neutron."

    I don't think you read his comments properly, as they were supporting your side not mine. He was saying that (current) atom lasers don't have enough KE to overcome Coulombic repulsion.

    But I'll disagree with him, as I don't feel there is any fundamental barrier to making the BEC move fast enough for nuclear collision purposes.

    The whole point of my suggestion of atom lasers for nuclear reactions, is to increase the probability of collisions, rather than depending on the stochastics of mean free path.
    Here you would create an anti-free path, by seeking to overlap 2 mutually opposing trajectories as fully as possible.
    And this is where the coherence of the colliding atoms comes in, because superpositioning them will give them less quantum jiggle, and synchronizing their jiggle with those of the matter they're colliding with would increase the chances of collision further still.

    If you are walking down the hallway while weaving left and right, and I am walking down the hallway while weaving left and right, then if our weaving motions are synchronized we can ensure the chances of our colliding with each other.
    Last edited: Jul 16, 2007
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook