Why does light diffract into only seven colours?

[sophiecentaur]

Isn't the multicolored visual effect produced by refraction?

It needs dispersion to produce a spectrum i.e. differential refraction.

Refraction / dispersion takes place at entry and exit. I seem to remember school experiments with glass prisms which showed that the greatest dispersion occurred with a symmetrical path through the prism ( i_in = r_out and i_out = r_in), on account of Snell's Law; or is that just rosy tinted prisms / specs?
The internal reflection is just an added complication which sends the light back towards an observer with their back to the Sun.

 

[hutchphd]

Thanks to @sophiecentaur and @sysprog for elucidating. The dispersion of color is caused by differential refraction of the frequencies of light at interfaces. For the raindrop the surfaces are curved and there can be several reflections internally, so the result is more complicated. But the physics is not fundamentally different.
Of course rainbows have more serendipity.

 

[DaveC426913]

Hence PANTONE came into being. Not cheap but it's a standard that gives a pretty good match between identical colours on different materials and surface textures.

Ah, but in my example, I neither had - nor gave you - a PANTONE chip.
An even if I had, you'd still have to recreate it perfectly consistently in your 10,000 gallons of paint.

Granted, PANTONE is one way of doing it. But they couldn't have done it without first quantifying their own colours.

 

[phinds]

Ah, but in my example, I neither had - nor gave you - a PANTONE chip.
An even if I had, you'd still have to recreate it perfectly consistently in your 10,000 gallons of paint.

Which is actually quite doable, apparently, since part of each of the Pantone colors is a specification for the appropriate ink mix. That's the only way that magazines can reliably and consistently reproduce specific Pantone colors. Without that capability the Pantone colors would be useless and they make a lot of money because they aren't useless.

Their testing facility (for batch consistency) is very impressive.

I can't remember whether Pantone colors actually apply to paint (I think they are not). They were designed for printers ink used in magazines.

 

[JeffJo]

If you don't think that a raindrop is a (spheroid) prism, please, in simple terms, say why not.

From Wolfram's Mathworld: "A general prism is a polyhedron possessing two congruent polygonal faces and with all remaining faces parallelograms (Kern and Bland 1948, p. 28; left figure)."

• I don't want to be as pedantic as some of these responses. This was about color, and why you should say "spectrum" and not "rainbow," but I keep being asked to explain what a rainbow is when a demonstration of "not a spectrum" should have been enough.
• I agree that you can get your meaning across by calling it a prism, but only when that shares the important properties. Here, the important property is not that it deflects light of different colors, but how it deflects them differently. A sphere does it differently that the prism as Wolfram defines it.

So, a sphere is not a prism. A prism has flat faces that deflect all of the (parallel) light of a single color the same way. That way you can reduce the effect to one dimension; a single point of entry, or entry along a line where all the light deflects the same way. A curved surface produces deflections that vary across the surface in a way that has to be accounted for. In a prism, you need first-year geometry to see where the light goes. In a raindrop, you either have to plot it or use calculus.

If you think that it is a (spheroid) prism, but "not how it's depicted", then please explain,

Pick your favorite explanation of rainbows. Most are based on a diagram like these:

And compare it to:

In most rainbow explanations, it is clearly implied - and sometime stated - that the rainbow is caused by light from a single white ray separating into colors like what happens in that prism. Or that each color emerges with a single angle of deflection. Since both statementgs about rainbows are wrong, they never explain why they think red light comes out only that one angle. As I keep repeating, that is not the case, and how these angles differ explains why a rainbow is not a spectrum.

Also please explain, in simple terms, how the multicolored visual effect is "not due to what people think of as the prism effect". Isn't the multicolored visual effect produced by refraction? And isn't it refraction that produces what people think of as the prism effect?

All that light coming out at the same angle is what I called "the prism effect." See the diagrams above.

I tried to explain this "prism effect" is what happens in Newton's experiment, the one he used to characterize colors, and is supposed to be the subject of this discussion. But a rainbow is not caused by all of the light of a single color emerging at the same angle. As I have said several times. That's why a rainbow is different than a spectrum.

Refraction / dispersion takes place at entry and exit.

<Sigh.> Dispersion is not synonymous with refraction, or separation. It is caused by refraction, and may or may not involve separation. Depending on what you think is separating.

Dispersion means that the path traveled by light, which was once independent of wavelength, must be treated as a function of wavelength. And no, I don't have a reference that uses that exact definition. Ignore that I said it if you don't like it, as it really isn't that important to why a rainbow is different than a spectrum. I only mention it because it when we treat similar, but different, concepts as equivalent that issues like spectrum-or-rainbow arise.

The internal reflection is just an added complication which sends the light back towards an observer with their back to the Sun.

The internal reflection is what creates the minimum deflection that makes rainbows. (Not to be confused with the "maximum deviation" sometimes used with prisms as you vary the angle of incidence.)

 

[sysprog]

I keep being asked to explain what a rainbow is when a demonstration of "not a spectrum" should have been enough.

I didn't ask you to explain what a rainbow is. I did ask you to, if you don't think that a raindrop is a (spheroid) prism, please, in simple terms, say why not. Your answer was essentially that it's not a prism because it's not geometrically a polyhedral prism. That difference was acknowledged parenthetically, so your answer that merely elaborates on that already acknowledged difference is unresponsive.

[*]I agree that you can get your meaning across by calling it a prism, but only when that shares the important properties.

Then the term 'spheroidal prism' is adequate to get the meaning across, and to acknowledge differences in properties.

Here, the important property is not that it deflects light of different colors, but how it deflects them differently. A sphere does it differently that the prism as Wolfram defines it.

The difference between a spherical or spheroid prism, and a polyhedral 'proper' prism, is acknowledged by the use of the modifying adjective.

So, a sphere is not a prism.

The term 'spherical prism' or 'spheroid prism' means a spherical or spheroidal object which produces a prismatic optical effect. It's clearly not referential specifically to a polyhedral prism. It's easy enough to understand that.

As I keep repeating, that is not the case, and how these angles differ explains why a rainbow is not a spectrum.

The term 'rainbow' designates the physical phenomenon by which spectral colors are displayed; it's not as precise in separating them as a glass prism is, but that doesn't mean that people are wrong when they say that a rainbow manifests the spectrum of visible light, or when they say that a raindrop is a spheroidal prism that produces a circular spectrum. That's part of a reasonable explanation for why, from the ground, the rainbow looks like a multlcolored semicircular arch.

All that light coming out at the same angle is what I called "the prism effect."

Maybe so, but you made reference to "what people call the prism effect", and that isn't restricted to all the light (for a given spectral frequency) coming out at the same angle, as individual light frequencies would in an abstractly perfect prism $-$ the effect by which white light is separated visibly into its consituent frequencies is reasonably called the prism effect, because the prism is the model object for production of that effect.

 

[sophiecentaur]

I agree that you can get your meaning across by calling it a prism, but only when that shares the important properties. Here, the important property is not that it deflects light of different colors, but how it deflects them differently. A sphere does it differently that the prism as Wolfram defines it.

Actually, the same laws of reflection and refraction apply to any shape of surface because they relate to the Normal. The fact that a raindrop has a curve makes no difference to what goes on; lenses and curved mirrors are designed on the basis of Ray Optics, which breaks any surface into elemental facets.

When right angled prisms are used in optical instruments, care is taken to ensure the light strikes input and output faces Normally, with the specific intention of eliminating dispersion but that is only an example of different behaviour from a raindrop.

An appropriate (shallow) three sided prism can be used to incorporate internal reflection and dispersion on the way in and out. Much the same effect as in a raindrop and with the same angles involved..

 

[hutchphd]

<Sigh.> Dispersion is not synonymous with refraction, or separation. It is caused by refraction, and may or may not involve separation. Depending on what you think is separating.

As with many foolishly contentious arguments this one now wallows in a sea of semantic purity devoid of Physics. I will now take a short swim myself.
To disperse means to "distribute or spread over a wide area" and dispersion is the act of so doing.
The classic Newton sunlight-through-the-window-shade experiment causes a projection (real image) of this color-dispersed light while the rainbow provides a virtual image of the color-dispersed light. In each case the separartion is caused by the same physics: the frequency dependence of the speed of light. This is in contradistinction to a diffraction grating whch really is different.
Whether you wish to call this the "prism effect" is a matter of personal preference.

Late to the party, I know. But in skimming the thread I didn't see the right answer, and did see much that is wrong. (And speaks poorly of Newton.)

Thanks for joining the party, but it is getting late..

 

[Mondayman]

My old coworker was color blind. Manufactuers would send furniture with wrong labels on occasion, so if he happened to deal with it, the customer would receive the wrong coloured furniture. To make matters worse, it always happened to somebody who was from out of town.

 

[hutchphd]

At one of my happiest jobs, the other physicst was red green colorblind and this was occasionally startling. I vividly remember Rick wandering into my office with two "gelatin" filters, one magenta and one forest green, to ask me which was which! He could not see any difference!

 

[sysprog]

Refraction / dispersion takes place at entry and exit.

<Sigh.> Dispersion is not synonymous with refraction, or separation.

It is not reasonable to interpret that sentence from @sophiecentaur as saying that dispersion is synonymous with refraction.

 

[hutchphd]

Yes. As a semantic issue I always interpret "/" to be shorthand for "and/or" and so this is exactly correct.