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

Infinite Positions in QM

  1. Feb 6, 2012 #1
    Last night I started reading the section of Brian Cox's "The Quantum Universe" where they discuss how to calculate the probability of a particle being found at a particular position.

    What he has stated first is if we have an initial position, at any later time we can find the particle at any spot in the universe. Okay, all good so far.

    An example is given where we roughly know where the particle is initially, and we want to calculate the probability of it being at spot X in the future. All possible routes from where it could be initially are calculated producing a final probability for X. Now they're saying because of interference, most of the routes cancel each other out (at least, I get that impression). But in the end, the conclusion is it "effectively has no chance of being found at X".

    Effectively no chance? No probability at all, or a small probability? In one breath you say a particle can hop to every other position in the universe in an instance (even if its in a superposition, saying it can be anywhere would seem to me that it does NOT have probability 0 of being at particular points. When you say something is in a superposition of state A and state B and state C, you wouldn't say its in a superposition of all three if state C had no chance of occuring).

    Would it be correct from the get-go we need to consider the particle to be in every possible position in the universe to calculate the probability we find it at position X?

    In all honesty reading this section has made me more confused.
  2. jcsd
  3. Feb 6, 2012 #2


    Staff: Mentor

    What I think you are talking about is the sum over histories approach and why most paths cancel. The answer I would give depends on how much math you know but since I cant really explain it without a bit of math I will give the answer with the least amount.

    Each path is represented by a complex number Ce^if - f is called the action. Both C and f depend on position and time. You may remember f can be interpreted as an angle and if it is rotated by 180 degrees then the number is the negative of what it was before. Now if f is a very large number then if f actually varies depending on the path a very close path can be found such that f changes by 180 degrees and since it is the negative of the other path they cancel. This will always happen with one exception - if for that close path f did not change - in which case instead of cancelling it will add. The paths where f does not change are called the paths of least action and, for all practical purposes, are the only ones that exist. This explains a very important principle of classical physics called the principle of least action and that principle alone, believe it or not is all that is required to derive all of classical mechanics.

    Check out:

    If you are interested in the technicalities of how Classical Mechanics is derived this way the book to get is Landau Mechanics:

    'If physicists could weep, they would weep over this book. The book is devastingly brief whilst deriving, in its few pages, all the great results of classical mechanics. Results that in other books take take up many more pages. I first came across Landau's mechanics many years ago as a brash undergrad. My prof at the time had given me this book but warned me that it's the kind of book that ages like wine. I've read this book several times since and I have found that indeed, each time is more rewarding than the last.'

    Many people feel like that after reading the book - it had a profound effect on me and from what I can gather many others that have read it.

    Last edited: Feb 6, 2012
  4. Feb 6, 2012 #3

    Math is not reality, it's just a tool that for some reason is successful in mimicking reality to a certain level of accuracy. WRT that, the above explanation is neither successful in addressing the question(as posed) nor complete. AFAICS, the question is more geared towards the interpretation side of things.
  5. Feb 6, 2012 #4

    The inability to know a position of a free particle anywhere in the universe follows from the wave nature of 'particles' and the HUP as well(this cannot be interpreted, as you well know, and is taken to be fundamental). And if it's really fundamental(we have been wrong before), it says a lot about realism.
    Last edited: Feb 6, 2012
  6. Feb 6, 2012 #5


    Staff: Mentor

    Of course math is not 'reality' - it however can be, and in this case is, a model of 'reality' (whatever that is) with strong experimental support. Next thing you will be claiming is the phasor diagrams engineers use to describe electric circuits cant possibly work because its just math.

    Now precisely what don't you understand about what I wrote? Not a philosophical objection but the nitty gritty and nuts and bolts of it?

  7. Feb 6, 2012 #6

    No, you fail to understand that things in reality don't work because the math says so(math has been adjusted to fit observations as much as possible and as close as possible, not the other way around). It would be a misconception and certainly not a mainstream view to claim otherwise. Moreover, you failed to demonstrate how the 'interpretation' of phasor diagrams is representative or relevant to interpreting the formalism of quantum mechanics and why you think such an analogy would be useful.

    Do 'particles' follow each path that is represented by a complex number as the 'nuts and bolts' of mathematics imply considering that you claimed of mathematics - "in this case is, a model of 'reality'" You know this? Really?

    You said "

    whereas the OP asked if it is possible in practice to

    Comparing the bolded text, do you still find your answer satisfactory?
    Last edited: Feb 6, 2012
  8. Feb 6, 2012 #7


    Staff: Mentor

    No one I am aware of claims that - I certainly do not. The claim is models (and QM is a model) that have been found to be correct by experiment to a high degree of accuracy have logical deductions that that have the same level of confidence as the experiments that support the model. It's the same thing that for example allows us to predict how a spinning top behaves from the mathematics of classical mechanics.

    Phasors were not mentioned as having anything to do with QM - it was mentioned as another example of a mathematical model that is used extensively and no one doubts its applicability.

    I think you need to think more about what a mathematical model is:
    http://www.math.ualberta.ca/~devries/erc2001/slides.pdf [Broken]

    Yes indeed. You do understand in QM states are modelled by complex numbers and that key fact is the reason for interference and that most paths are cancelled by a very close path? This is combined with the fact that the f I mentioned usually has a factor 1/hbar in front of it I omitted for simplicity and since hbar is very small it makes it very large. Have you read Feynmans QED - The Strange Theory Of Light And Matter? The turning arrow he talks about is a complex number which has both length and an angle and explains what I said in a slightly different way. If you haven't here is a link to the actual lectures:

    Last edited by a moderator: May 5, 2017
  9. Feb 6, 2012 #8
    Nope, HUP says that since we measure it and have completely determined its position, its momentum is completely undetermined and thus it could exist at any energy level. We can't base the information of the future on the present. All we can do is gather data and look for patterns, such as that "we keep seeing it appear in this general area, and it has a few specific energies , so we can say that when energy=x1, x2, x3, it will generally appear over in these regions".

    "routs" don't really cancel each other out in just a plain standing wave, what happens is just the position itself is undetermined, kind of like how "1/0" is undefined or doesn't have a specific outcome since it can technically yield any number as a result.

    At large distances probability comes so close to zero that scientists just consider it 0, just like how the mass of an electron in chemistry is so small we don't consider it. We still only have general areas of probability of measuring a particle at a given energy even if it can appear somewhere else in the universe. Matter isn't completely a particle or completely a wave, you could sort of say that it exists more where its more probable and exists less where its less probable.
  10. Feb 6, 2012 #9

    What you're referring to is "almost never", which is a similar concept to "almost surely".

    Edit: oh yeah, I forgot to mention: an event with probability 0 CAN happen. For example, in the darts scenario on the unit square (the example provided by wiki), the probability of hitting the point (1/2,1/2) = 0, BUT it is still POSSIBLE!
  11. Feb 7, 2012 #10

    Wouldn't it have been much easier for you to supply a straightforward answer to the question as posed in the OP, instead of beating about the bush:

    Since you believe the formalism is true because it's empirically verified, the ONLY obvious answer is that according to the SE the answer is that there is a non-zero chance that the particle can be at X,Y,Z at time T somewhere in the universe, whereby specifying that the time T may come after 2 billion years or more(or less).

    How is the following quote from the OP not a philosophical question?(did you read the question at all?)

    Absolutely NOT. You are again confusing models for reality and deluding yourself that everything flows out of the formalism, whereas the only thing that can be said categorically is that interference is CAUSED by the wave nature of 'particles'. I don't find this point particularly hard to grasp. I would guess the reason the comedy arises in the first place is that an instrumentalist attempts to answer an otherwise obvious philosophical question and tries to sweep the philosophical implications under the rag with comments like - for all practical purposes. You could have simply said - qm is not classical mechanics and things don't behave the same way as they tend to do macroscopically.
    Last edited: Feb 7, 2012
  12. Feb 7, 2012 #11


    Staff: Mentor

    Wouldn't it be easier for you to make an attempt to understand it instead of confusing your inability to do so with the belief I did not supply an answer?

    Figures. How about actually reading even a basic textbook?

    Last edited: Feb 7, 2012
  13. Feb 7, 2012 #12

    What you supplied was not an answer to the question but to a hypothetical question that was not(asked about practical purposes). Why is that not fairly obvious by now?

    Setting aside the adhominem attack, my point was that the probability of finding a particle at any point (its "probability density") was related to the square of the height of the probability wave at that point. That quickly vanishing amplitude extends throughout space(with non-zero values) which answers the opening post.
  14. Feb 7, 2012 #13


    Staff: Mentor

    Maybe because the confusion lies with you rather than what I wrote?

    What kind of number is a probability amplitude in QM?

    I was going to leave it at that but to try and stop this going on longer than necessary here is the answer:
    'In quantum mechanics, a probability amplitude is a complex number whose modulus squared represents a probability or probability density. For example, if the probability amplitude of a quantum state is α, the probability of measuring that state is | α | 2. The values taken by a normalised wave function ψ at each point x are probability amplitudes, since |ψ(x)|2 gives the probability density at position x.'

    It is the fact it is a complex number that is responsible for interference effects and for nearby paths canceling - exactly as my original post explained. Now I again ask what about do you not understand? Is it because you don't know what a complex number is? No shame in that. If you don't then have a look at the Feynman lectures I gave a link to - he explains it without using that concept.

    Maybe if I explain it another way it will be clearer. The probability of finding a particle at any point is the square of a complex number called the probability amplitude. In the sum over history approach that number is the contribution of the sum over all possible paths ie you need to add up all the complex numbers that each path gives at that point. Now consider the contribution of some path - it turns out to have the form Ce^if (which is of course the form of any complex number) where f has the form of g/hbar - the fact hbar is small is very important. Since hbar is very small this means f will be a very large number. Now consider a close path such that f changes by pi. Since this is such a large number C will not change much for such a close path so that you end up with minus Ce^if, which when added to the original path results in a big fat zero - ie the contribution of that path is cancelled. There is only one case where this will not happen - where f does not change when a small change in the path is made - in which case it will add instead of cancel. This means only the paths where f does not vary over the path are left as contributing to the probability of finding a particle - and it usually turns out to be just one path where that happens. And that is why in most cases the particle follows a distinct path - it is the path where when you make a small change in the path the probability amplitude does not change.

    Last edited: Feb 7, 2012
  15. Feb 7, 2012 #14
    If you graph the function of a probability wave, there is a horizontal asymtote at x=0 which means the probability never reaches 0 but only comes infinitely close to it.
  16. Feb 12, 2012 #15
    Please please. No need to have a go at each other. I have found each and every post helpful. Thank you all for the answers. :)
  17. Feb 15, 2012 #16
    So to clarify:

    Every quantum system in the universe, evolving according to Schrodinger's equation, has a non-zero probability of being anywhere else in the universe at a later time? Would that even include a particle being used in a lab experiment?
  18. Feb 15, 2012 #17
    You can't predict where a particle will show up based on present information. Or I guess you can say that you can, but even in a very isolated system you still have like a 50% chance of being wrong.
    If you look at some graphs of probability waves, there are some very specific places where the wave intersects the x-axis, but after those x values it just asymtotes. Of course as time goes on, that probability changes, so the places that use to intersect the x axis are no longer there and there is never a consistant place that the probability is 0.
  19. Feb 15, 2012 #18
    I'm not sure if that answers my question. Of course we cannot predict where a particle will be at a later time with certainty. We can, however, predict probabilities for every possible result from every future possible experiment conceivable based on a starting wave function.
  20. Feb 16, 2012 #19
    I don't think we can do that either, unless Douglas Adams invented the infinite improbability drive in his slot in the cemetery.
  21. Feb 16, 2012 #20
    From point A to point B, a particle can take an infinite amount of paths to get to the next point. What paths are possible? Ones that don't currently have other systems in them?
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook