Duality Of Light Explored By Revisiting The Double-Slit Experiment

We’ve all seen recreations of the famous double-slit experiment, which showed that light can behave both as a wave and as a particle. Or rather, it’s likely that what we’ve seen is the results of the double-slit experiment, that barcode-looking pattern of light and dark stripes, accompanied by some handwaving about classical versus quantum mechanics. But if you’ve got 20 minutes to invest, this video of the whole double-slit experiment cuts through the handwaving and opens your eyes to the quantum world.

For anyone unfamiliar with the double-slit experiment,  [Huygens Optics] actually doesn’t spend that much time explaining the background. Our explainer does a great job on the topic, but suffice it to say that when coherent light passes through two closely spaced, extremely fine openings, a characteristic pattern of alternating light and dark bands can be observed. On the one hand, this demonstrates the wave nature of light, just as waves on the ocean or sound waves interfere constructively and destructively. On the other hand, the varying intensity across the interference pattern suggests a particle nature to light.

To resolve this conundrum, [Huygens] jumps right into the experiment, which he claims can be done with simple, easily sourced equipment. This is belied a little by the fact that he used photolithography to create his slits, but it should still be possible to reproduce with slits made in more traditional ways. The most fascinating bit of this for us was the demonstration of single-photon self-interference using nothing but neutral density filters and a CCD camera. The explanation that follows of how it can be that a single photon can pass through both slits at the same time is one of the most approachable expositions on quantum mechanics we’ve ever heard.

[Huygens Optics] has done some really fascinating stuff lately, from variable profile mirrors to precision spirit levels. This one, though, really helped scratch our quantum itch.

30 thoughts on “Duality Of Light Explored By Revisiting The Double-Slit Experiment

    1. Agreed, same here. Not too-high-class school in Germany, early 1980s. I don’t remember the material but would suspect aluminum.
      As far as I understand, the double slit experiment is much more interesting when looked at (pun intended) from the perspective of “how does the wave function work, does it collapse or is a collapse unnecessary, e.g. in a multi-worlds setup?”. The experiment in this scenario isn’t about wave-versus-particle, it’s about probabilities and single photons.

      However, I didn’t watch/read/listen to the articles linked, so maybe this modern use of the experiment is being explained there. The “classic” view of it, to me, is really early-school-material and not too exciting to begin with.

    2. Not to be a d*ck but why even post your comment. Nowhere did anyone say it was hard to observe, rather that it is a well done video explaining the physics and especially the trickier quantum mechanical aspects of it. I would argue that the single photon experiment IS traditionally hard to setup without expensive light sources and detectors, all moot now with the availability of cheap lasers and cmos sensors.

      1. I’d venture to say, most of these people would never bother to question anything, let alone design an experiment to test it. The guy in this video is nothing short of genius, and I don’t say that lightly. Not only did he create a extremely precise experiment (ffs human hair) with off the shelf parts, he also had the ability to understand what the results were telling him, and was able to convey it better than any explanation I’ve personally ever heard for the dual slit, and I’ve heard them all.

    3. Getting an interference pattern with bright laser is easy, but measuring the probability distribution when only a single photon arrives at a time is more difficult. Even more difficult is detecting the single photons individually, so that you can conclusively say that they are discrete particles, but simultaneously exhibit the probability distribution.

    1. For many modern image sensors the difference between one intensity level and the next is just a handful of photons (especially for 10 or 12 bit AD conversion and high gain it may be pushed down to single photons). Separating the signal photons from the noise floor is the challenge in most applications. Here this is done by averaging many, many images.

    2. Electron-multiplying CCD sensors have been around and available for 15+ years. They function like and array of solid-state photomultipliers, and are designed to unambiguously detect single optical photons.

      Marketing blurb here: https://andor.oxinst.com/learning/view/article/electron-multiplying-ccd-cameras

      They are expensive, their SNR is poor, dynamic range not great, and their quantum efficiency is not very good.

      Unless you really, really need single-photon detection you are almost always better off using a modern CCD or even a CMOS detector. Modern detector readout noise is approaching (and sometimes better) than a single electron, and quantum efficiencies can exceed 90%. In other words, pretty much every photon arriving gets detected in a modern detector, but you don’t have to pay the price of an EM-CCD’s poor SNR and dynamic range (and cost).

    3. It’s not a single photon, as he explains in the video. It’s a beam with an average of one photon in the beam at any time. There are still millions of photons which accumulate over even very short periods of time, but on average there’s only one photon in-flight at any given time.

    1. First, it isn’t every photon – there are many lost just getting through the optics. But the bottom line is that the energy of a single photon is known, and a light source can be attenuated to the point where photons are being emitted at a known rate, like one per several seconds. You can then measure the response of sensors, while turning this light source off and on, to demonstrate that there is a corresponding increase in the sensor output. If this is the case, the infrequency of the photons guarantees that only single photons are hitting the sensor. There is still a certain probability that occasionally two or three photons arrive close enough in time (the “married” photons you propose) that there could be a cumulative effect, but this would be orders of magnitude less.

      I was playing with a Tektronix 7104A oscilloscope once, back in the 80s, which used a microchannel plate as an electron multiplier, to dramatically brighten the image for very fast single-shot events. I cranked it up to its maximum sweep speed, put a viewing hood over it, and triggered it. The displayed trace was quite granular – it appeared to be composed of overlapping dots. Was I seeing individual electrons hitting the MCP, or the “clumping” of bunches of electrons? If I knew the CRT’s cathode current, I probably could have determined this, because I could have calculated the number of electrons emitted in each picosecond, and counted the average number of bright spots in the display per cm, but alas, I didn’t.

      1. You are sort of seeing individual electrons. I built a microchannel plate amplify while working on my PhD and the way they work is that a single or few particles strike the plate and it amplifies them into a burst. So, you are seeing a burst of many electrons that have originated from one or only a few electrons that activate the plate.

        1. Thank you. That’s what I thought. The dots I was seeing were overlapping by quite a bit, so I wasn’t seeing clearly distinct dots, but a texture, if you will, that looks like a grainy photograph. Which could be explained by either individual electrons whose “bursts” cover enough area that they overlap, or statistical highlights in the rain of electrons, where each of these highlights is caused by multiple electron strikes. As I said, if I knew the beam current, I probably could have determined which of these was the case. It was still a pretty cool experiment, though, in that I was clearly seeing the EFFECTS of individual electrons.

    2. Interestingly, this was figured out in the 1920’s before modern electronics.

      You can shine a beam of light on a thermometer and measure the increase of temperature, and from that deduce the power (energy per second). (Glossing over some details, but accurate measurement of temperature isn’t too hard.)

      From the power and knowing the wavelength, you can calculate the total number of photons per second.

      Make the light parallel and put the beam through a pinhole and you will reduce the power that doesn’t go through the pinhole, which also proportionally reduces the number of photons.

      Hold a piece of glass over a candle to deposit carbon and block out some of the light, and then measure how much power gets through using the thermometer technique above. Make several of these, and label how much light gets through each. (This is how the original experiment did it 100 years ago.)

      With your reduced photons through the hole, place a selection of filters in the beam and calculate the number of photons that get through by multiplying the blocking factor of each.

      You can get the photon beam down to 1 photon a second on average this way, and have single photons to use for your physics experiment.

      This was basically what kicked off quantum mechanics in the last century: Einstein proposed that light comes in discrete units, so people started generating single photons to see what sorts of properties they had, such as with beam-splitter mirrors and double slits.

      And the result, that single a photon appears to take both paths, was a complete surprise to everyone.

      1. Interestingly the double-slit experiment with feeble light (although probably not down to single-photon level) was done in 1909, about a century after Young carried out his original double-slit experiment. British physicist G.I. Taylor performed his experiment in a very dark room and using a photographic plate as the screen in which he showed that even the feeblest light source at his disposal – equivalent to “a candle burning at a distance slightly exceeding a mile” – would lead to interference fringes.

    3. There is such a thing as “photon bunching”, however, this happens with negligible frequency, so the probability of there being two photons in flight simultaneously within the apparatus is thus negligible. All the interference effects observed can be attributed to the weird (quantum) behavior of individual photons, traversing through the double-slit one at a time.

      For purists, an entangled photon source can be used in advanced single-photon experiments because the downconversion process intrinsically produces two beams of individual photons that verify each other’s existence. Photons reaching the experiment’s detector are counted only if they coincide with the detection of its partner photon. This eliminates the possibility of “photon bunching” which may affect single-photon sources based on strongly attenuated laser beams.

      This is an area in which I’ve done quite a bit of experimental work, and I believe that for this purpose photon bunching can be safely ignored. Jeroen’s work (AKA YouTube’s Huygens Optics) is TRULY well done.

      1. I think as far as “photon bunching” goes, it’s mostly irrelevant, because even if you did occasionally get multiple photons in a bunch, there would be some single photons that would hypothetically not interfere.

        If that were the case, the distribution would change, and you would get more proportionally more photons hitting the dark areas at lower light levels. so with a beam like this where there’s probably only a single photon in the beam the majority of the time, you would expect a more uniform distribution with only subtle interference bands from the small portion of photons that interfere.

        I think the more difficult thing to think about, is if it takes multiple photons to interfere, how close together do they have to be (in either time, or distance) to interfere with each other.

        1. Photons are bosons: they can’t “interfere with each other”.
          The interference pattern is solely from photons that interfere with themselves, bunching or not.

          (next level, to forestall the pedants: In non-linear media, you can get multi-photon processes due to the medium properties, but that’s not photons interfering with each other)

  1. HaD story intro’s are like tabloid headlines “We’ve all seen recreations of the famous double-slit experiment,” followed shortly by “For anyone unfamiliar with the double-slit experiment”.

    I got stuck in that loop and my computer nearly exploded.

  2. His youtube channel is pretty amazing, best on youtube for anything optics related by far. If you were ever curious about how lenses are cnc ground and lapped, his channel covers it all. One of my favorite channels, I highly recommend a look to the curious

  3. These oversimplified explinations of quantum mechanics are the reason why there is so much misunderstanding around it. The experiment here explains only the classical, wave, nature of light and trying to claim that this explains the quantum (ie finite and descritized) nature of light is like explaining that a baloon floats because it has negative mass. If you define the problem statement right sure it ‘makes sense’ but it blatently wrong and gives people the wrong idea about density/boyancy/etc.

    What is interesting about the quantum nature of light is explicitly that, when you start to observe the quantized nature of photons, that light does not behave in the way you would expect based on this double slit experiment! The statistics start to get wonky in a way that is inconsistant with the behavior you observe with a simple ‘bright light’ setup like this one–see things like bells inequality for where things really go off the rails.

    1. Seems to me like the single-photon part of the experiment was a pretty good demonstration of the quantum nature of light. I don’t think one photon going through two slits at the same time when they’re more that the size of the photon apart is something that can be explained by classical mechanics, but I could be wrong.

  4. “On the other hand, the varying intensity across the interference pattern suggests a particle nature to light.”
    This is one interpretation, that light are both particles and waves and that the dots on the screen represent each photon.
    However, a second interpretation is that it is the interaction of the light with matter that results in an isolated interaction. In other words, it is a property of the matter on the screen molecules, which requires a multiple of the Planck constant in order to bump an electron to a higher energy level, that creates the appearance that the light is a particle. This interpretation allows the light to travel as waves yet interact as quanta.

  5. Photons going everywhere at once?

    Well, they are radio waves, after all. Try stopping radio waves going everywhere at once – you’ll have a hard time doing that!

    I guess people think of a visible light photon being small, because a single atom can “receive” it, but actually, that’s just the “antenna” as it were.

    That was my “take away” idea from this video, that the photons spread out infinitely unless constrained. Now my previous preconceptions have been neutralised, single photon interference does make much more sense.

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.