3
$\begingroup$

Towards the end of the following video https://www.youtube.com/watch?v=GzbKb59my3U the double slit experiment is executed with 'single' photons and it is shown how the interference pattern emerges as the aggregation of registrations of single photons behind the slit.

How slow can this experiment be done and still show the interference pattern?

In other words: at what interval between photons reaching the detector will the interference pattern not be visible any more in the detection data? us? ns? seconds? years?

Is this effect dependent on the interval at all? If so how?

Improve this question
$\endgroup$
5
  • 1
    $\begingroup$ if you mean lowering the rate of incident photons, the only difference is that there will be a higher time-interval between detection events and therefore you will have to wait a longer time to see the interference pattern emerge $\endgroup$
    – glS
    Commented Mar 21, 2015 at 18:50
  • 1
    $\begingroup$ @glance is this something actually experimentally verified? $\endgroup$
    – pointernil
    Commented Mar 21, 2015 at 22:46
  • 1
    $\begingroup$ the rate can be infinitely low, so you cannot in principle verify it for any rate. It is certainly verified that the interference of photons with themselves is independent of the emission rate, for most reasonable experimental regimes. I myself routinely see it in the lab with single photons emitted through SPDC at various intensities of the laser pump. There is currently no reason to think that this would break in other regimes, and it would most probably disprove the whole quantum mechanical formalism. $\endgroup$
    – glS
    Commented Mar 22, 2015 at 11:35
  • $\begingroup$ @glanc thanks! "it would most probably disprove the whole quantum mechanical formalism quantum mechanical formalism". I sure hope enough people are trying that already ;) For science! $\endgroup$
    – pointernil
    Commented Mar 22, 2015 at 16:23
  • 1
    $\begingroup$ I don't know what your background is here, but the belief of most non-scientists that somehow in science things are taken for granted without experimental evidence cannot be more false. In particular in the case of quantum mechanics you can be sure that every (a lot of) vailable alternatives have been tried. Every bit of quantum mechanics has been and currently is challenged, sometimes to an almost ridicolous level (the current issue of loopholes comes to my mind), because nothing is taken for granted. $\endgroup$
    – glS
    Commented Mar 22, 2015 at 19:29

2 Answers 2

Reset to default
4
$\begingroup$

Fundamentally, you can slow down double-slit experiments as much as you like but a version of the Heisenberg principle will not allow you to put definite "gaps" between the detection events or to be sure to produce a photon on demand, at the right frequency etc. It is commonly stated that the particle must be interacting with itself in a double slit experiment, but when you take a look exactly at these uncertainties of the sparseness of particles in the beam along with the fact that there is no clear picture of the single-particle interaction, you have to take the textbook story about "particle self-interaction" with a grain of salt.

However, as technology progresses we actually have working single-photon sources which are commonly used in both science (quantum optics) and even industry (development of imaging devices).

As mentioned, these sources cannot surpass a certain amount of indeterminism. The quantum picture of a continuous beam of photons as a stationary wave-function is only replaced by a picture of separate gaussian-like pulses in the wave function into which a single photon might or might not squeeze in. But you might actually squeeze two photons into the pulse. Or three, or none. The design of a single-photon device is such that the probability of such events is minimized but it is there and the minimization comes at cost of another uncertainty in the produced photon. I.e., when you get down to it, all the quantum weirdness stays in it's full extent even for single-photon devices.

Improve this answer
$\endgroup$
3
  • 1
    $\begingroup$ Hold on a moment and think about the interaction between the slits edges and the photon. Is there a field interaction between the surface electrons of the edges and the photon? Has this field to be quantized? Could the fringes be the manifestation of this field? It can be proved or falsificated using different electric potential for the slit screen and observe the fringes width. $\endgroup$
    – HolgerFiedler
    Commented Mar 21, 2015 at 21:21
  • $\begingroup$ @Void any hints towards experiments trying to minimizing the number of photons used in the double slit experiment and still being able to show interference in a statistically significant way? What is the current lower bound? $\endgroup$
    – pointernil
    Commented Mar 21, 2015 at 22:58
  • 2
    $\begingroup$ @pointernil Well, you need a large number of photons to show interference in a statistically significant way, that's the point. But the single-photon sources such as a quantum dot can be pumped to emit in single independent events with arbitrary time intervals in-between. However, the emission event may produce more than one photon at a time. For the current state of the art single-photon emitters this happens in about 10-20% of cases. Once again, this statistic is obtained through detection events and you cannot jump to conclusions about how much beable photons interact in the apparatus. $\endgroup$
    – Void
    Commented Mar 22, 2015 at 12:04
2
$\begingroup$

When we look at individual photons, the experiment is about how the individual photons interact with themselves. Since the photon is interfering with itself and not with other photons in the experiment, the interval between the photons does not matter.

In a real experiment, however, you have to deal with the fact that your screen is not perfectly isolated from the environment. So, you have some rate of false readings that should be equally distributed on the screen. As a result, you have to be emitting photons at such a rate that your experimental data is not overwhelmed by the false readings. This is an effective limit on how slowly we can emit photons, but it will be dependent upon the experimental set up and the rate of false readings instead of a manifestation of the underlying quantum mechanical principles.

Improve this answer
$\endgroup$
1
  • 1
    $\begingroup$ Hmm, along which theoretical lines can we actually rule out that f.e. (just a very layman idea!!!) the 'space' between the slits and the detector is contributing to the 'accumulation' of the interference pattern? $\endgroup$
    – pointernil
    Commented Mar 21, 2015 at 22:58

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.