A question about the speed of Light in different media....

[Kaneki123]

Okay...The speed of light is affected by density of the medium in which it propagates...It is more in rarer medium and lesser in denser medium...Hence we can conclude that a medium affects the speed of light...My question is that why this ''affect'' is not ''continuous'', like why the speed of light is not constantly changing in the same medium?...Any help is appreciated...

 

[DrStupid]

Okay...The speed of light is affected by density of the medium in which it propagates...It is more in rarer medium and lesser in denser medium...

Counter example: Lead glass has a density of 3500-4800 kg/m³ and a refractive index of around 1.6. Dimond has a density of 3100-3500 kg/m³ and a refractive index of 2.4.

I think what you mean is optical density.

Hence we can conclude that a medium affects the speed of light...My question is that why this ''affect'' is not ''continuous'', like why the speed of light is not constantly changing in the same medium?

If the speed of light is determined by a property of the medium, then why should the speed of light change when this property remains constant?

 

[Dale]

My question is that why this ''affect'' is not ''continuous'', like why the speed of light is not constantly changing in the same medium?

It can change within the same medium. That is the basis for mirages and stars twinkling

 

[sophiecentaur]

It can change within the same medium. That is the basis for mirages and stars twinkling

The density can change with temperature and that can alter the speed within the same 'medium'.

 

[Buckleymanor]

Someone will correct me if I am wrong that the density does not have to change within a medium to alter the speed of light.
White light shone through a prism (glass) will form a rainbow of colours each colour will have a different frequency than another and will travel at a different speed.

 

[sophiecentaur]

to alter the speed of light.

You are 'not wrong' but quoting the speed assumes a particular frequency. You can't change the frequency, once the wave is launched.

 

[DrDu]

I think the frequency of light propagating in a gas like air decreases constantly due e.g. to Raman scattering, however, this effect is very small.
Nevertheless it leads theoretically to a continuous change of speed.

 

[Kaneki123]

Counter example: Lead glass has a density of 3500-4800 kg/m³ and a refractive index of around 1.6. Dimond has a density of 3100-3500 kg/m³ and a refractive index of 2.4.

I think what you mean is optical density.
If the speed of light is determined by a property of the medium, then why should the speed of light change when this property remains constant?

The speed of light in glass is approximately 200,000 kilometres (120,000 mi) /s, i.e, the speed of light changes as soon as the light enters from a different medium to glass...My question is that why, after this ''change'', the speed of light suddenly becomes constant for glass..., like why there is no constantly decreasing speed of light as the light propagates through glass?

 

[DrDu]

The speed of light in glass is approximately 200,000 kilometres (120,000 mi) /s, i.e, the speed of light changes as soon as the light enters from a different medium to glass...My question is that why, after this ''change'', the speed of light suddenly becomes constant for glass..., like why there is no constantly decreasing speed of light as the light propagates through glass?

What is propagating in glass is not light but so called excitons. I.e. the energy of the light is partially stored in atomic or molecular exited states and only a fraction of the time in photons moving at the vacuum speed of light. Hence on average, light is not as fast in a medium as in vacuum.

 

[jbriggs444]

The speed of light in glass is approximately 200,000 kilometres (120,000 mi) /s, i.e, the speed of light changes as soon as the light enters from a different medium to glass...My question is that why, after this ''change'', the speed of light suddenly becomes constant for glass..., like why there is no constantly decreasing speed of light as the light propagates through glass?

If I understand the concern, you envision light particles as little projectiles that slow down when they encounter dense objects -- like a bullet that slows down continuously as it passes through a tank of water or a slab of ballistic gelatin. But, as @DrDu points out, light particles are not little bullets.

Nor are they waves, exactly. But their propagation matches wave behavior.

 

[Kaneki123]

If I understand the concern, you envision light particles as little projectiles that slow down when they encounter dense objects -- like a bullet that slows down continuously as it passes through a tank of water or a slab of ballistic gelatin. But, as @DrDu points out, light particles are not little bullets.

Nor are they waves, exactly. But their propagation matches wave behavior.

If their propagation matches wave behaviour, should'nt there be some damping when light propagates through a medium?And if there is damping, then why is speed of light constant?

 

[jbriggs444]

If their propagation matches wave behaviour, should'nt there be some damping when light propagates through a medium?

Wave speed is independent of wave amplitude.

And if there is damping, then why is speed of light constant?

The above is only a heuristic answer, of course. Since light is only wave-like. But experiment shows that light does move at a measurable, constant speed in a uniform medium.

 

[sophiecentaur]

if there is damping, then why is speed of light constant

What I am getting from your posts is that you just don't believe what PF is telling you; your intuition is winning over the consensus. It may be a good idea to start considering that the PF message is probably (more) correct and what you need is arguments to support the idea that you could be wrong (?).
Why would you think that damping should slow up a wave? It will slow up a car but that's not the same thing. There are other ways of losing energy than losing kinetic energy (which, I think, why you want the wave to slow up). If the medium is isotropic (I'm talking in classical terms and let's get that sorted out first) then the dissipated energy can also cause the amplitude to drop on the way through without changing speed.
Also, consider this situation. A train of pulses of light are launched into glass block. You suggest that they should slow up on the way through. Imagine another beam of pulses, introduced into the block, half way along (the two beams come from a single modulated laser with a beam splitter). With your model, the second string of pulses wouldn't have started slowing down until it enters the block so they would arrive at the other end going faster than the original pulse stream. The frequency of the two streams of pulses would be different even though they have been traveling side by side through half of the block. The would arrive at a different rate. Have you ever heard of an experiment in which that happens?

 

[DrDu]

I already pointed out that incoherent scattering will lead to a change of frequency and in general also of speed. So the question is rather why this process is so inefficient with light.

 

[sophiecentaur]

I already pointed out that incoherent scattering will lead to a change of frequency and in general also of speed. So the question is rather why this process is so inefficient with light.

Has the conventional aspect of this question been resolved, though? I think we need one step at a time.

 

[Buckleymanor]

You are 'not wrong' but quoting the speed assumes a particular frequency. You can't change the frequency, once the wave is launched.

Why not.

 

[phinds]

Someone will correct me if I am wrong that the density does not have to change within a medium to alter the speed of light.
White light shone through a prism (glass) will form a rainbow of colours each colour will have a different frequency than another and will travel at a different speed.

The colors of light that exit a prism certainly do have different frequencies (the very definition of color in one way) but certainly do NOT have different speeds.

 

[sophiecentaur]

Why not.

It's not 'never' the case but a classical model is based on having an array of oscillators consisting of masses linked with springs. The force on one of the oscillators will be due to the displacement of the spring (or springs) and its period has to be the same as the displacements of the previous oscillators. So the only oscillation that can be induced in one oscillation has to have the same frequency as that of the previous oscillator. If not, the relative phases would march along until the motions were in the opposite direction ( 180 degree phase difference) and back again to in-phase. Even if there is some friction involved, this can't happen and any energy loss has to be by dissipation of energy and not by a change in frequency. There will be a steady phase lag as you look along the chain of oscillators , which represents the delay due to the speed of propagation.
Conventional wave theory talks in terms of phase continuity across a boundary. The phase step from incident wave to transmitted wave that can occur is only possible because of a reflected wave being present.
Once that is OK then you can discuss more advanced models - such as what DrDu mentioned - but it's not a good idea to jump into harder models until the basics are working for you. And . . . . . the effect of 'tired photons, as they go through a medium is very small.

 

[sophiecentaur]

do NOT have different speeds.

. . .once they exit the prism into space / air. (to emhasise)
The fact that they speed up again can give problems.

 

[DrDu]

It's not 'never' the case but a classical model is based on having an array of oscillators consisting of masses linked with springs. The force on one of the oscillators will be due to the displacement of the spring (or springs) and its period has to be the same as the displacements of the previous oscillators. So the only oscillation that can be induced in one oscillation has to have the same frequency as that of the previous oscillator. If not, the relative phases would march along until the motions were in the opposite direction ( 180 degree phase difference) and back again to in-phase. Even if there is some friction involved, this can't happen and any energy loss has to be by dissipation of energy and not by a change in frequency. There will be a steady phase lag as you look along the chain of oscillators , which represents the delay due to the speed of propagation.
Conventional wave theory talks in terms of phase continuity across a boundary. The phase step from incident wave to transmitted wave that can occur is only possible because of a reflected wave being present.
Once that is OK then you can discuss more advanced models - such as what DrDu mentioned - but it's not a good idea to jump into harder models until the basics are working for you. And . . . . . the effect of 'tired photons, as they go through a medium is very small.

Yes, but inelastic scatteting is exactly what slows down a classical barticle in a medium (aka friction), so you have to discuss its effect on light propagation to understand this question.

 

[sophiecentaur]

Yes, but inelastic scatteting is exactly what slows down a classical barticle in a medium (aka friction), so you have to discuss its effect on light propagation to understand this question.

What classical particles are involved in the 'latest' (i.e. not classical particulate) theory of light? The particles involved in a sustained mechanical wave will not 'slow down' in a way that will alter the frequency of the wave. Photon scattering is not the same as mechanically linked and vibrating particles. Why are you trying to rush through this, missing out steps?
Exploring light in terms of a wave motion and getting fully conversant with that model will avoid some of the obvious misconceptions that can arise from treating light as a wave 'wrongly'.
If you try a classical model of light in involving particles (and a classical model can't do both models at once) then you get nonsense if you try to explain refraction or diffraction (which are the dominant phenomena which need explaining) . If you want to go further and introduce your scattering phenomenon, you have to take a big step and get into QM. That's fine, of course, but isn't it a risk if the classical model isn't thoroughly sorted out? PF is full of threads from people who want to talk advanced Physics without sorting out the basics and it always causes tears before bedtime.

 

[sophiecentaur]

Yes, but inelastic scatteting is exactly what slows down a classical barticle in a medium (aka friction), so you have to discuss its effect on light propagation to understand this question.

Friction doesn't cause the frequency to change, tho'. I think you should be separating what happens to classical particles and to
quantum particles. I think that the distinction is very important, so much so that they should have two different words. Feynman'a fault.

 

[DrDu]

Friction doesn't cause the frequency to change, tho'. I

I found a classical paper by Chandrasekar from 1948 where the the softening of gamma radiation due to multiple Compton scattering is calculated:
http://rspa.royalsocietypublishing.org/content/192/1031/508

I would paraphrase this exactly as a frequency change due to friction.

 

[sophiecentaur]

I found a classical paper by Chandrasekar from 1948 where the the softening of gamma radiation due to multiple Compton scattering is calculated:

Sorry but I think it will cost me money to read that. Is the only ref that you can quote, sixty years old? Chandrasekar is a very creditable source but are you sure that you are interpreting his message as he intended? Did no one else have the same idea and publish?
In any case, Compton Scattering involves Photons and they are not part of classical wave theory. I realize fully that you can come to all sorts of conclusions by involving QM but the friction I have been discussing can only involve drag on the vibrations in a (classical) system of linked mechanical (and massive) components. In a classical forced oscillation with damping, how can you get a frequency change? I don't see what you are trying to prove here. Notice that the thread has a B classification.

 

[DrDu]

Personally, I tend to use the simplest modelling level to explain phenomena that is available. A particulate theory is time honoured since Newton. Yes, you can derive a classical particle model as a limit from full QED as you can derive a classical field theory. Of course you can derive frequency shifts from statistical fluctuations with a certain power spectrum of the dielectric constant to obtain a classical wave theoretic description: If the dielectric constant or polarisability has a fluctuation which oscillates with frequency $\Omega$, this will modulate the incident light of frequency $\omega$ and yield the Stokes and anti-Stokes sidebands with frequency $\omega \pm \Omega$. I think I learned this in a physics lab (for chemists!) in my first term at university where there was an experiment on Raman scattering, so I would call it B-level.

The short answer to the clever questions of Kaneki123 is that you can get increasing frequency shift from incoherent scattering for light as you get slowing of motion for a particle in a medium. If the medium is dispersive, this leads also to a change in speed. The scattering of light from electrons hardly changes the energy of the photons, i.e. frequency change is small due to the small momentum of photons as compared to electrons. In a qualitative sense light behaves here as if consisting of very "light" corpuscules as compared to electrons from which they scatter. Hence this is a rather weak effect, however, it becomes important e. g. in glass fibers where optical path length is very large and Brillouin and Raman scattering represent mayor sources of loss.

 

[sophiecentaur]

I don't know why we are having this conversation any more. The thread is classified with an I. Whether or not it is appropriate for the OP, we don't know but we should surely answer and discuss at that (I) level. I don't know at what stage you were educated about QM in any meaningful way but I have not come across many A level students who would be able to deal with Schrödinger and it's certainly not (as of a few years ago, at least - and certainly not in either of our educations) part of what they get taught.
The primary phenomena that are observable (by us and the 'ancients') are diffraction based and very easily explained by looking at the wave nature of light. Of course, wave theory is not enough to explain frequency reduction and your more advanced ideas need to be brought in.
If you want a higher level of conversation then why not either ask the OP directly if the change of agenda is OK or else start a new thread with the appropriate level. Alternatively if you have a way that 'explains' simple optics in terms of particles that doesn't involve statistics, QM, or harder Maths and which can yield the results of Young's Slits and how a lens works then go ahead.

Personally, I tend to use the simplest modelling level to explain phenomena that is available.

It all depends how "simple" you find the names and ideas that you include in that first paragraph in your post. Can you rely on a member who posts an 'I' level question finding it that simple? You'd wonder why Huygens and Maxwell ever got involved with waves, if particles make it so simple.

 

[DrDu]

Nevertheless, I think I learned an interesting piece of physics from this question and though it is not classical wave mechanics, you don't need to solve Schroedinger equations.
Here is little back of the enelope estimation, dropping all numerical constants, of the length a photon can travel in matter until it has lost
all its energy due to Compton scattering:
The density of electrons in ordinary matter is about 1 electron per cube of the Bohr radius $a_0$, i.e. $V=a_0^3$.
The scattering cross section for Compton scattering in the low energy range is about $\sigma=r_e^2$ with $r_e$ being the classical electron radius.
The change in wavelength per scattering event is about $\lambda_e$, the Compton wavelength of the electron.
Now I can ask how many scattering events it takes for a photon with wavelength $\lambda$ to loose all its energy.
This is $n=\lambda/\lambda_e$.
The mean free path length between scattering events is $V/\sigma$, or the total path length till all energy is lost is
$l=\lambda V/(\sigma \lambda_e)$. Now the three length scales are all proportional to each other $r_e=\alpha \lambda_e=\alpha^2 a_0$ with $\alpha=1/137$ being the fine structure constant.
Hence
$l=\lambda /\alpha^5$.
For X rays this makes only some cm, but for visible light some dekameters.
This is an asymptotic result which should hold when wavelength is so high that electrons can be considered approximately free.
This is certainly reasonable for x-rays and maybe for UV light in metals above the plasma frequency.
For transparent media like glass it is probably still too small because the momentum transfer is suppressed by the electrons being bound to atoms. I would estimate this effect introducing another factor $M/m_e$ where M is the typical mass of the atoms. So light can travel several hundred kilometers in glass until it has been scattered completely by incoherent processes.

 

[sophiecentaur]

Nevertheless, I think I learned an interesting piece of physics

Me too.

 

[cranksci]

"The speed of light in glass is approximately 200,000 kilometres (120,000 mi) /s, i.e, the speed of light changes as soon as the light enters from a different medium to glass...My question is that why, after this ''change'', the speed of light suddenly becomes constant for glass..., like why there is no constantly decreasing speed of light as the light propagates through glass?"

I understand your question. The term "speed of light" is somewhat of a misnomer, because what we are really talking about here is the rate of the exchange of information between adjacent participants in the propagating medium, which is a constant value that we know to occur over a distance of 186,000 miles in one second in the medium of air, wherein that determination is made. Denser mediums have more participants in the chain of information exchange over the same distance, therefore light will take longer to cross the denser medium for the same time frame, and when returned to the medium that is air, that distance increases with the decrease in density of the medium, and the apparent "speed of light" immediately returns to "normal".

 

[nasu]

The phase velocity changes as well when the wave (not necessarily EM wave) enters a different medium. The problem is not to distinguish between phase, group and signal velocities. They all change in a different medium.
I think the confusion comes from the fact that we use "velocity" for two slightly different things. The velocity of a moving particle is related to change in position of a material object over time. In the case of a wave we have a similar velocity for the particles of the medium, in the case of a mechanical wave. But what we call wave velocity does not describe quite the same thing.
No particle moves with that velocity. For phase velocity is the phase, an abstract concept, that change position. The concepts of accelerating and decelerating the wave does not apply to this type of velocity. The fact that there is a tendency to "jump" to photons before understanding properly the classic wave makes the confusion more likely.

I did not see any question about accelerating and decelerating sound waves when they go from air to water and vice-versa. Would that mean that the things are pretty clear for sound?