A view of the double-slit experiment

[Qhmu]

Greetings. My name is Antti, I'm from Finland. My scientific background consists mostly if not only from watching youtube videos about science and Googling things that I'm curious about. So i registered on this forum to ask questions from the more educated people.

So here's my understanding as to why light seems to behave like a wave in the double-slit experiment:

In case the picture for some reason isn't visible: In my understanding, light bounces from the insides of the slits onto the screen, creating a cool pattern that seems like the output of waves.

Could this be correct?

 

[ZapperZ]

Could this be correct?

No, it can't be.

You need to figure out what the pattern will look like for ONE slit. In your "model", adding another slit simply means that the pattern is repeated, but shifted by a distance. This is definitely not what is going on with the double-slit pattern.

Furthermore, you need to be aware that this double-slit pattern occurs not just for light, or even for electrons being shot at some opening. It also occurs for currents and other quantum particles that can take two different paths. In those cases, there are no "slits" and no bouncing off the walls of the slits. This means that this interference effects is a more general phenomenon than what you think.

Zz.

 

[Henryk]

In my understanding, light bounces from the insides of the slits onto the screen

Actually, the light doesn't just 'bounce' from the insides of the slits, instead, it is exhibiting diffraction.
Diffraction is one characteristic of waves that let waves 'go around obstacles'. Another aspect of diffraction is spreading out of the wave after it passes through an aperture or opening. If a light passes through a large opening, it just propagates along straight lines. However, if an opening is comparable to the wavelength of the light, the light starts spreading after it passes through an opening. There is a video that shows clearly diffraction of the waves on water surface, take a look.

So, in the double slit experiment we have two wave phenomena: diffraction and interference. The slits are narrow (relative to the wavelength of light) and that causes the light from each slit spread laterally out due to diffraction.
Then, the resulting waves from both slits interfere to produce the periodic pattern on the screen.

 

[Qhmu]

Thanks for the answers. I took a look at the video, what are the "spiky" things in the insides of the opening? Sorry, these kind of questions probably belong to the homework-section or yahoo :D

Edit: Oops, i forgot to "model" the single slit pattern

 

[PeroK]

Thanks for the answers. I took a look at the video, what are the "spiky" things in the insides of the opening? Sorry, these kind of questions probably belong to the homework-section or yahoo :D

Edit: Oops, i forgot to "model" the single slit pattern

Try the Khan Academy videos on diffraction, single slit and double slit, as a good start.

You should also learn about the Huygens principle.

 

[Qhmu]

Hello, i created a simulation of what i meant by bouncing creating diffraction, here's a short 30 second clip, you can skip to ~10 second mark to see better:



The "waves" in this simulation consist of circles that don't collide with each other shooting in all directions at the same speed from different points and not losing energy on collision.

Isn't this what happens when waves get to an opening? Btw, sorry about the bad quality.

 

[ZapperZ]

Hello, i created a simulation of what i meant by bouncing creating diffraction, here's a short 30 second clip, you can skip to ~10 second mark to see better:



The "waves" in this simulation consist of circles that don't collide with each other shooting in all directions at the same speed from different points and not losing energy on collision.

Isn't this what happens when waves get to an opening? Btw, sorry about the bad quality.

First of all, this isn't a "simulation". It is more of a guess work. To simulate means that you have (i) a mathematical form that you are using for modeling and (ii) you show how your result actually MATCHES the observation that we know of.

It appears that you are not aware of the patterns for both diffraction and interference. Because of that, I'm going to post one of my ppt lecture notes on here that shows the relationship between single-slit diffraction and double-slit interference.

Look very closely at what is going on. There are nodes and antinodes in the pattern. And you appear to not understand my earlier statement that in your model, all you get are DOUBLE the pattern you get for a single slit, with a small shift. This is NOT what you get here. The single slit pattern MODULATES the interference pattern from the double slit.

This is not A² + B². This is more like (A + B)², where there is a cross-term in the description. And not only that, your model does even have ANY explanation or simulation to show the presence of nodes and antinodes.

I can show many more of these, especially on what happens when the slit width is changed, and when the distance between the slits is changed. These are all described mathematically.

Again, please remember that physics just doesn't say "what goes up, must come down". It must also say "when and where it comes down". It must have a quantitative part that matches observation. This is where the mathematical description comes in.

Zz.

 

[Qhmu]

Thanks for the reply, i have to admit that it's all guesswork and attempting to make it overly simple for myself. I'll have to look into this closer.

Peace.

 

[ZapperZ]

Thanks for the reply, i have to admit that it's all guesswork and attempting to make it overly simple for myself. I'll have to look into this closer.

Peace.

I know that you are trying to learn. However, you also need to make sure that before you attempt to provide a description or explanation of something, you need to know what that THING is as much as possible in the first place. You are trying to offer an alternative explanation to a phenomenon that is very well known in physics. And you should also keep in mind that you are presenting this to an audience that includes physicists. So trying to do this for something that you don't fully understand (the full diffraction and interference patterns) to a bunch of people who know it way more than you do, it doesn't look very good.

I've offered this advice for many new members here who think they have a possible explanation for something. Make sure you understand that thing first. In your case, you should have asked if you have understood all the different types of diffraction and interference patterns first before attempting to "model" them. After all, does it make any sense to you to try to model an observation when you don't even know the full extent of that observation?

In other words, try and learn how to walk first before attempting the 100 m sprint.

Zz.

 

[Qhmu]

What i don't understand is that if i put two mirrors close together facing each other and shine a flashlight through the gap, it creates the same "interference" or overlapping reflections pattern.

If i start shrinking the mirrors, at which point the surfaces of the mirrors stop being surfaces?

 

[Ibix]

Generally when all sizes (both the mirrors and the spaces between them) are much larger than the wavelength of the light then you can use ray optics as an approximation. Which is what I think you are trying to do. The closer together the mirrors are the less accurate that is, because you're simplifying the wave nature of light. It becomes obviously wrong when there is a length scale that gets close to the wavelength of light.

 

[Qhmu]

But at which point in size a surface ceases to be a surface? The slit-screen is so thin that the walls of the slits aren't surfaces?
I mean, how a pinhole camera works if not like that?

 

[Ibix]

What do you mean by when a surface is not a surface?

A pinhole camera can be described by ray optics, yes. But a typical pinhole is a fraction of a millimetre across. A wavelength of light is a fraction of a micrometre. A pinhole is enormous compared to the wavelength of light.

 

[Qhmu]

I don't know enough of physics to prove or disprove any theory, it just looks to me that the double slit-laser pattern is just a selfie of the laser pointer in different angles and places, what happens to the pattern if you turn the laser around it's axis slightly?

Please forgive my lack of knowledge and words

 

[Ibix]

The laser's beam profile doesn't have much effect on the pattern. The slits are much narrower than the laser beam so the slits are more or less uniformly illuminated. The pattern depends on the slit width and spacing. It is true that the maxima look a bit like the laser beam profile, but that's just in a "bright in the middle and dimmer at the edge" kind of way. Careful plotting of the intensity will show you otherwise - typically the maxima look something like $\sin^2x$ and laser beams like $e^{-x^2}$.

Rotating the laser beam won't usually do anything because the beam profiles are typically circularly symmetric (there are exceptions).

I still don't understand what you mean by a surface not being a surface.

Edit: by the way, does everyone else see two post #10s (one by Qhmu, one my first post) in this thread, or is it just me?

 

[weirdoguy]

does everyone else see two post #10s

Yep.

 

[ZapperZ]

I don't know enough of physics to prove or disprove any theory, it just looks to me that the double slit-laser pattern is just a selfie of the laser pointer in different angles and places, what happens to the pattern if you turn the laser around it's axis slightly?

Please forgive my lack of knowledge and words

But you should have enough knowledge to read our responses on why we have disproved YOUR "model", and I'm using the word "model" very loosely here because in scientific terms, your scenario does not qualify to be called a model, because you have no mathematical description.

But something from another thread just reminded me of another evidence on why your scenario doesn't work. It is the diffraction that we get from an edge. Edge diffraction means that one can see light "bending" around a corner and produces this diffraction pattern in a region where it is in the "shadow" of the light source. Using your scenario of light simply bouncing of surfaces, such observation should not be possible, and yet, there it is!

Zz.

 

[Qhmu]

The laser's beam profile doesn't have much effect on the pattern. The slits are much narrower than the laser beam so the slits are more or less uniformly illuminated. The pattern depends on the slit width and spacing. It is true that the maxima look a bit like the laser beam profile, but that's just in a "bright in the middle and dimmer at the edge" kind of way. Careful plotting of the intensity will show you otherwise - typically the maxima look something like $\sin^2x$ and laser beams like $e^{-x^2}$.

Rotating the laser beam won't usually do anything because the beam profiles are typically circularly symmetric (there are exceptions).

I still don't understand what you mean by a surface not being a surface.Edit: by the way, does everyone else see two post #10s (one by Qhmu, one my first post) in this thread, or is it just me?

Thank you, i meant the slit plate's thickness, not slit width, a hole or a slit has surfaces, i don't understand why a laser wouldn't illuminate the surfaces and give off reflections if the slit width isn't huge? Yep, i also see two post #10s

Edit: also a reflection should get distorted if the direction of the reflection and the surface the reflection travels to arent aligned perfectly

But you should have enough knowledge to read our responses on why we have disproved YOUR "model", and I'm using the word "model" very loosely here because in scientific terms, your scenario does not qualify to be called a model, because you have no mathematical description.

But something from another thread just reminded me of another evidence on why your scenario doesn't work. It is the diffraction that we get from an edge. Edge diffraction means that one can see light "bending" around a corner and produces this diffraction pattern in a region where it is in the "shadow" of the light source. Using your scenario of light simply bouncing of surfaces, such observation should not be possible, and yet, there it is!

Zz.

Thanks, i'll give it a read. By the way, you are the only one here calling those "models".

 

[Ibix]

Thank you, i meant the slit plate's thickness, not slit width, a hole or a slit has surfaces, i don't understand why a laser wouldn't illuminate the surfaces and give off reflections if the slit width isn't huge?

Because that's not how waves behave when interacting with objects that are a similar size to them.

You can only use the "light goes in a straight line and bounces off things" model (i.e. ray optics) when everything is much bigger than the light wavelength. It's just an approximation - one that works very well in everyday life, but is wrong if pushed too far. You are pushing it too far.

If you want to understand diffraction you need to let go of ray optics and study light as a wave. Huygens' Principle would be a good place to start. You'll need to be able to do integration for anything more than idealised cases.

 

[Qhmu]

Because that's not how waves behave when interacting with objects that are a similar size to them.

You can only use the "light goes in a straight line and bounces off things" model (i.e. ray optics) when everything is much bigger than the light wavelength. It's just an approximation - one that works very well in everyday life, but is wrong if pushed too far. You are pushing it too far.

If you want to understand diffraction you need to let go of ray optics and study light as a wave. Huygens' Principle would be a good place to start. You'll need to be able to do integration for anything more than idealised cases.

Yeah, i was just about to let this thread sink, i just somehow thought it was ridiculous that light would somehow start behaving differently in smaller scale. Please excuse my stubbornness.

 

[Ibix]

Yeah, i was just about to let this thread sink, i just somehow thought it was ridiculous that light would somehow start behaving differently in smaller scale. Please excuse my stubbornness.

It doesn't really behave differently at a small scale - you can always use the wave model of light. But if everything is much larger than the wavelength of light it turns out that this is a mathematically complex and time consuming way of getting a result that you probably won't be able to distinguish from the result of ray optics. So we use ray optics - but always keep one eye on the sizes of things in case diffraction and interference might be important and we need to use a more complex model.

This is very common throughout physics. For example NASA mostly use Newtonian gravity for sending space probes around the system because errors from design tolerances, micrometeorite impacts and the like are much, much larger than those from not using Einstein's theory of gravity. So why waste computer time? But the GPS satellite clocks are adjusted for gravitational time dilation (a prediction of Einstein's theory) because it's important that the clocks be accurate to that precision.

 

[Qhmu]

It doesn't really behave differently at a small scale - you can always use the wave model of light. But if everything is much larger than the wavelength of light it turns out that this is a mathematically complex and time consuming way of getting a result that you probably won't be able to distinguish from the result of ray optics. So we use ray optics - but always keep one eye on the sizes of things in case diffraction and interference might be important and we need to use a more complex model.

This is very common throughout physics. For example NASA mostly use Newtonian gravity for sending space probes around the system because errors from design tolerances, micrometeorite impacts and the like are much, much larger than those from not using Einstein's theory of gravity. So why waste computer time? But the GPS satellite clocks are adjusted for gravitational time dilation (a prediction of Einstein's theory) because it's important that the clocks be accurate to that precision.

To be honest, the way i tried to understand how entanglement or superposition works led me to this kind of thinking:

What i mean is, if i stand in a room with a strobe lighting in front of a mirror facing the mirror, i know i am facing the mirror, so the mirror image must be facing the opposite direction, if i turn left, i know that the mirror image must turn right(relative to the mirror). And the light coming "from me" to the mirror and back isn't really at two places at once, if i were to measure the light i would find it either in me or the mirror, but never in both places simultaneously?

 

[Ibix]

This is all classical optics. Quantum theory is not needed and trying to deduce stuff about quantum theory from thinking about this will leave you with nonsense. For example, the light is never in the mirror, so your analogy makes no sense.

If you want to understand entanglement and superposition then you need a textbook. Griffiths is the one I usually see recommended here.

 

[Qhmu]

This is all classical optics. Quantum theory is not needed and trying to deduce stuff about quantum theory from thinking about this will leave you with nonsense. For example, the light is never in the mirror, so your analogy makes no sense.

If you want to understand entanglement and superposition then you need a textbook. Griffiths is the one I usually see recommended here.

I see

Edit: I see no point in continuing this discussion any further, it revealed an interesting pattern of human behaviour and was very eye opening, thank you all taking part in this.

 

[James Briggs]

I may have missed an answer but most people are missing the point. IT DOESN'T BOUNCE with one slit. There is a reason why they use to slits. Take one electron at a time. With one slit it goes right through. In other words with one slit it acts like a particle. If it bounced around it would do it with one slit. With two slits it doesn't bounce around but acts like a wave. Use a detector to show which slit it s using and it acts like a particle. Take the detector way and it acts like a wave.

 

[Ibix]

I may have missed an answer but most people are missing the point. IT DOESN'T BOUNCE with one slit.

It doesn't bounce, full stop. That kind of model cannot explain diffraction, which is a phenomenon related to waves.

Take one electron at a time. With one slit it goes right through. In other words with one slit it acts like a particle.

This is not correct - google for "single slit diffraction".

Use a detector to show which slit it s using and it acts like a particle. Take the detector way and it acts like a wave.

This thread has largely been about classical electromagnetism where this kind of experiment isn't possible. That said, it is true that single photon/single electron patterns differ depending on whether you track which slit the particle goes through or not. However, you don't get "particle" or "wave" behaviour. You either get a double-slit diffraction pattern (no detector) or two overlapping single-slit patterns (with detector).

I suspect that a smart experimentalist will arrange the experiment so that the single-slit patterns are near-field and look very like a naive "no diffraction" model while the double-slit pattern is in the far-field, so that the distinction between one result and the other is very clear. It's still not strictly correct to say that there is no diffraction in the single-slit case.

Edit: there's no interference between the slits in the with-detector case. But that's a different thing.

 

[Qhmu]

I don't understand. A wave seems travel or "expand" in all directions at the same speed, why would it stop doing that inside any slit or hole, hitting the walls leading to "flipping" the wave and changing it's shape?

 

[PeroK]

I may have missed an answer but most people are missing the point. IT DOESN'T BOUNCE with one slit. There is a reason why they use to slits. Take one electron at a time. With one slit it goes right through. In other words with one slit it acts like a particle. If it bounced around it would do it with one slit. With two slits it doesn't bounce around but acts like a wave. Use a detector to show which slit it s using and it acts like a particle. Take the detector way and it acts like a wave.

An electron diffracts through a single slit. Try this:

http://www.cornell.edu/video/richard-feynman-messenger-lecture-6-probability-uncertainty-quantum-mechanical-view-nature

Skip to 24:30 if you are short of time.

 

[Ibix]

I don't understand. A wave seems travel or "expand" in all directions at the same speed, why would it stop doing that inside any slit or hole, hitting the walls leading to "flipping" the wave and changing it's shape?

Because they're waves. They don't bounce round and bounce off each other like particles, they interact and interfere. Did you look up Huygens' principle?

 

[Qhmu]

Because they're waves. They don't bounce round and bounce off each other like particles, they interact and interfere. Did you look up Huygens' principle?

I did, but everyone is basically saying that a wave stops expanding inside an opening and doesn't collide with the "walls" at all?

A perfect plane wave should collide with more points on the walls than a circular shaped wave

What a wave is even made out of? Is water made of waves?

 

[PeroK]

What a wave is even made out of? Is water made of waves?

Water is made of water molecules. A water wave is the propagation of energy through the water. Try this:

https://www.khanacademy.org/science...ound/mechanical-waves/v/introduction-to-waves

 

[Qhmu]

Water is made of water molecules. A water wave is the propagation of energy through the water. Try this:

https://www.khanacademy.org/science...ound/mechanical-waves/v/introduction-to-waves

But the waves themselves are still just water molecules?

 

[PeroK]

But the waves themselves are still just water molecules?

The waves are motion/energy of the water molecules. You might as well ask whether the speed/kinetic energy of a car is made of car parts.

 

[Qhmu]

I might as well go to a christian forum and try to tell them how walking on water was an optical illusion, and they would start citing the bible.

Again, everyone is basically saying that a wave stops expanding/starts behaving differently the moment it gets inside an opening and doesn't collide with the walls of said opening?

 

[PeroK]

I might as well go to a christian forum and try to tell them how walking on water was an optical illusion, and they would start citing the bible.

Again, everyone is basically saying that a wave stops expanding/starts behaving differently the moment it gets inside an opening and doesn't collide with the walls of said opening?

It's only you saying that, as far as I can see. If the walls of the opening are thin, then they would have little effect. If, however, the opening was a tunnel, then the water would move constrained inside the tunnel until it emerged. This behaviour, however, is not really relevant to the diffraction, as that is what happens when the water emerges from the opening, not what happens when the water is constrained by the opening.