I have done many double and single slit experiments where I created clear interference patterns. But one of the amazing things about the double slit is that under certain conditions the wave properties of light go away and the particle properties of light are created. How can I create an experiment where the interference pattern is replaced with a pattern that shows the particle properties of light?


Actually the particle properties of light are there all along, and the pattern would be the same.

To see light as particles in the double-slit experiment, you have to shoot photons individually and see them form the interference pattern dot by dot.

What would make a different pattern though is a detector able to record which slit a photon uses. You would then not get any more interferences, just the sum of two single-slit diffraction patterns.

  • $\begingroup$ So the particle properties would be to watch the individual photons slowly make an interference pattern? Is there ever a configuration that would cause the photons to act like little bullets and simply make two dots on the screen that correspond to each slit? $\endgroup$ – Lambda Jan 11 '17 at 0:31
  • $\begingroup$ The particle property is whenever it is useful to talk about "photon". In the double-slit case, it appears when you reduce the intensity of light. But this does not change the interference pattern: the wave property is still present. That's what is called wave/particle duality. $\endgroup$ – Stéphane Rollandin Jan 11 '17 at 0:38
  • $\begingroup$ Also, photons can definitely not be seen as little bullets. They do not have well-defined trajectories. See What exactly is a photon?. $\endgroup$ – Stéphane Rollandin Jan 11 '17 at 0:43
  • $\begingroup$ Now about having dots corresponding to each slit: remember that the slits have a width allowing diffraction of the light, else you would not have the interference pattern in the first place. So the photons are being diffracted by each single slit too (in other words, if they could be seen as having a trajectory, it would not necessarily be a straight line). $\endgroup$ – Stéphane Rollandin Jan 11 '17 at 0:50
  • $\begingroup$ I guess I was just wrong in my assumption that at some point light would act like bullets. That's what I get for relying on YouTube as my main source. There a quite a few videos on there that show one line of impact that corresponds to each slit. $\endgroup$ – Lambda Jan 11 '17 at 0:59

You can detect the individual photons building up the interference pattern by using a digital camera to record the interference pattern. You then need to place the camera behind the slits and you need to make sure the photo-sites of the sensor are going to be hit no more than a few hundred times. This is because you need to use the highest ISO setting where the dynamic range will be the lowest.

If you use a 1 mw green laser and the entire interference patters covers all the, say, 20 megapixels, then there are about 140 million photons hitting each photo-site per second. If you expose for, say, 1/4000 seconds, you'll still have 35000 photons hitting each photo-site, which is too much. You can reduce the amount by using a neutral density filter. If you use one that reduces the light intensity by a factor of 1000, you'll have on average 35 photons per photo-site per exposure, allowing you to see the discrete steps in intensity at each pixel.

You'll also have to deal with noise and the fact that only a fraction of the photons hitting a photo-site will actually be detected, these factors depend on the quality of the camera.

  • $\begingroup$ I was under the impression that the particle characteristics of light would show two spots on the screen, as if little balls were thrown through the slits. Are you saying that will never happen? That the "particles" will always make an interference pattern? $\endgroup$ – Lambda Jan 11 '17 at 0:36
  • $\begingroup$ @Lambda photon particles do make the pattern. $\endgroup$ – Bill Alsept Jan 11 '17 at 23:02
  • $\begingroup$ Bill, please explain. $\endgroup$ – Lambda Jan 12 '17 at 0:24
  • $\begingroup$ @Lambda see my paper "single edge certainty" billalsept.com. It was published in this months edition of infinite energy magazine. $\endgroup$ – Bill Alsept Jan 12 '17 at 2:57

Sure, you need to detect which slit the photon goes through. In principle you could do it like this:

  1. Hang a plate from an apparatus that is robust against transferring momentum to it.
  2. Cut two slits into the plate that have splitting on the order of the photon wavelength to set up the regular double slit scenario.
  3. Design a very sensitive momentum sensor for the plate (for instance put reflecting mirrors on the ends and set up a laser interferometer)
  4. Shoot one photon at a time at the center of your plate, detect which way the plate is kicked due to conservation of momentum.
  • $\begingroup$ Your setup is very creative, but I have shone a bright laser directly onto an interferometer's mirror and even the full force of the laser did not move the interferometer's image. $\endgroup$ – Lambda Jan 11 '17 at 0:52
  • 1
    $\begingroup$ If you could detect such tiny impulses, they would tell you how the direction of each individual photon changed after passing through the opening, but knowing only how the direction changed is not going to tell you which slit a photons passed through. $\endgroup$ – Solomon Slow Jan 11 '17 at 2:16
  • $\begingroup$ @jameslarge If they go through the right slit, the mirror must move to the left since you are aiming at the center and visa versa. And Lambda, I'm not sure you totally understood the idea. $\endgroup$ – Bobak Hashemi Jan 11 '17 at 5:30
  • $\begingroup$ @BobakHashemi; Re, "If they go through the right slit, the mirror must move to the left." The only way there could be a force on the plate is if it changes the direction of a photon, and a single slit can deflect a photon in either direction en.wikipedia.org/wiki/Diffraction#Single-slit_diffraction $\endgroup$ – Solomon Slow Jan 11 '17 at 18:39

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