What is the relationship between the speed of light and the true vacuum?

[Crazymechanic]

Hi , well we all know that the speed of light in vacuum is measured at c and it slows down when it enters other mediums.
Now we always say speed of light is c in vacuum but hence I believe the speed is c there because there is nothing in the way that could slow light down like there is in other mediums like water , air , different types of gasses etc.
But we also know that there are different types of vacuums out there , some mediums are closer to a true vacuum and some are not , like there is a partial little vacuum in the intake manifold of a internal combustion engine but we wouldn't say that light would travel at c in there would we?

So the question then is , how "empty" the vacuum needs to be for light to travel at c in it ?

For example would the vacuum which is commonly used in a CRT tube is enough ?

 

[Drakkith]

I would guess that you need a perfect vacuum to get light to travel exactly at c over a significant distance.

Given the density of gas in a very very good vacuum is extremely low, my question is what effect would a single particle have on the expanding wavefront? Does it slow it down momentarily? How much of the wavefront does it affect?

 

[nsaspook]

Air is still very empty at typical RF (< 1 ghz) frequencies as the size of the molecules in air are tiny when compared to the space between them and the wavelenght of the EM waves. The EM field gradient across anyone molecule would be a very small fraction of the total field strength at any point in space so it's interaction with that energy would be very small if that molecule was isolated from others.

The best vacuum levels we see on normal measurement devices in the semiconductor industry might be in the e-12 Toor range (using Ion pumps) for a very good CD SEM at the field emission tip.

http://www.virginia.edu/ep/SurfaceScience/class2.html

UHV is typically considered to be below 10-9 Torr. Even at the best vacuum normally used in surface science experiments, ~10-11 Torr, there are still plenty of molecules in the gas phase, about 300 hundred thousand per cm3. In interstellar space, the density of molecules can be as low as a few per cm3, corresponding to about 10-16 Torr.

 

[Crazymechanic]

Hmm so , let me guess when we measured light here on Earth with our apparatus we didn't have a perfect vacuum not even close? Now I believe they came to the final speed taking inaccuracies into context or they just did a better job by calculating the speeds from cosmic intergalactic distances yet as the speed of c is also used to say the distances between cosmic objects they first had to know the speed precisely in order to tell how far something is.

 

[technician]

why not turn it on its head?
If you measure the speed of light in a region of space and get 2.99792458 x 10^8 m/s then that is a perfect vacuum.

 

[BruceW]

or, from an experimental viewpoint: if the pressure of your 'almost vacuum' is some value epsilon, then by decreasing epsilon, you should be able to get a value of the speed of light to be closer to c. (Until epsilon gets small enough that quantum effects come into consideration).

 

[Crazymechanic]

Well as technician said I believe the best measurement they have done is up in space.

 

[voko]

Now I believe they came to the final speed taking inaccuracies into context or they just did a better job by calculating the speeds from cosmic intergalactic distances yet as the speed of c is also used to say the distances between cosmic objects they first had to know the speed precisely in order to tell how far something is.

In the end they just decided that measuring the speed of light was not worth it, and just DEFINED it to be a particular number of meters per second. And then they defined one meter to be the length light propagates to with a particular period of time.

So now they don't measure the speed of light anymore; they measure the meter.

 

[jbriggs444]

In the end they just decided that measuring the speed of light was not worth it, and just DEFINED it to be a particular number of meters per second. And then they defined one meter to be the length light propagates to with a particular period of time.

My take is that they decided that measuring the speed of light was already being done so precisely, so accurately and was so easily reproducible and that measuring the length of the second was being done so accurately, so precisely and was so easily reproducible that one can do a better job establishing a standard of length based on a standard second and the speed of light than by depending on some carefully inscribed scratches on a standard hunk of metal stored in a laboratory.

That said, I expect that the techniques used to realize the current length standard probably have little to do with what one would ordinarily think of as measuring the speed of light.

 

[Drakkith]

Hmm so , let me guess when we measured light here on Earth with our apparatus we didn't have a perfect vacuum not even close? Now I believe they came to the final speed taking inaccuracies into context or they just did a better job by calculating the speeds from cosmic intergalactic distances yet as the speed of c is also used to say the distances between cosmic objects they first had to know the speed precisely in order to tell how far something is.

Here you go: http://en.wikipedia.org/wiki/Speed_of_light#Measurement

 

[voko]

My take is that they decided that measuring the speed of light was already being done so precisely, so accurately and was so easily reproducible

Not really. They understood that the speed of light was a fundamental constant. So instead of having two units defined independently and with some independent uncertainty, the left just one - the time unit - defined with an uncertainty, and the other one related to it exactly via the constant.

 

[Nugatory]

H
Now we always say speed of light is c in vacuum but hence I believe the speed is c there because there is nothing in the way that could slow light down like there is in other mediums like water , air , different types of gasses etc.

Be careful here. Light slowing down in a medium isn't like a car being slowed by air resistance or you finding that wading upstream against a current is more work than wading downstream with the current. Instead, the light is traveling at c through the vacuum between the particles of the medium; but when the light interacts with the particles it can be absorbed and reemitted after a time delay, reflected, bounced off in a different direction. It's not so much that the light is slowed because something is "in the way"; it's more that if you look closely enough it travels in fits and starts.

For example would the vacuum which is commonly used in a CRT tube is enough ?

How good are your measuring instruments? For any practical problem that I can think of offhand, a CRT tube is a good enough vacuum that you can treat it as perfect. For that matter, as long as we're working with light you don't need a vacuum at all; the speed of light in ordinary air is only about 100 km/sec less than it is in the hardest of hard vacuums. That's not a big number compared with 3x[10⁸ km/sec.

 

[BruceW]

Be careful here. Light slowing down in a medium isn't like a car being slowed by air resistance or you finding that wading upstream against a current is more work than wading downstream with the current. Instead, the light is traveling at c through the vacuum between the particles of the medium; but when the light interacts with the particles it can be absorbed and reemitted after a time delay, reflected, bounced off in a different direction. It's not so much that the light is slowed because something is "in the way"; it's more that if you look closely enough it travels in fits and starts.

(I think) You can either talk about a photon quasiparticle, or talk about actual photons, which are absorbed and emitted. I agree though, in either 'picture', we don't have actual photons traveling at some speed other than c.

 

[D H]

In the end they just decided that measuring the speed of light was not worth it, and just DEFINED it to be a particular number of meters per second.

That is not how things proceeded. It took 96 years of ever improving measurements of the invariance of the speed of light, and 78 years of experimental verifications of special relativity before the meter was redefined as the "length of the path traveled by light in a vacuum in 1⁄299,792,458 of a second".

Not worth it? Those ever improving measurements of the speed of light, along with those ever improving demonstrations that the speed of light is invariant were absolutely critical in making this redefinition acceptable to the international metrology community.

 

[voko]

@ D H

My remark was supposed to be somewhat humorous.

Anyway, I do think that "not worth it" captures the idea that with the invariance of the speed of light widely recognized as a fundamental principle of nature, measuring it really means measuring the units of time and/or distance, hence the re-definition.

 

[russ_watters]

I'd just say it a different way: the meter and C were redefined because it was realized that the speed of light is more accurately known and fundamental than the the meter, so the meter should be defined based on the speed of light and not the other way around.

Edit: voko beat me: that's close to my characterization.

 

[Dale]

So the question then is , how "empty" the vacuum needs to be for light to travel at c in it ?

For example would the vacuum which is commonly used in a CRT tube is enough ?

You should start with the section called "optical extinction" here
http://www.edu-observatory.org/physics-faq/Relativity/SR/experiments.html

It also has some original references.

 

[Crazymechanic]

Ok so basically we could say that light travels at c at any given moment and medium but hence we usually measure the distance that light has traveled in the macro level we say that light travels at c only at vacuum because at other mediums the absorption and re-emission of individual photons takes time so hence the light hits the measuring target later than if it would have been "free" on it's way all the time.

 

[technician]

I'd just say it a different way: the meter and C were redefined because it was realized that the speed of light is more accurately known and fundamental than the the meter, so the meter should be defined based on the speed of light and not the other way around.

Edit: voko beat me: that's close to my characterization.

So...am I correct in saying that now the speed of light is a defined quantity and that the metre is a derived quantity?
Or...is 'derived' the wrong word to use in relation to the metre.
Can you recommend any references in this area?I have just checked: 1 metre is defined as :
"length of the path traveled by light in a vacuum in 1⁄299,792,458 of a second".
So it is the metre that is defined

 

[Crazymechanic]

I think this argument is more about the sake of it not about some reality or some real quantity we use everyday.
For example , measuring one metre can vary depending on the man doing the measurement on the measuring tape on the object that is being measured etc.
Now on the other hand we have found out that light is fixed by nature at a given number on a given medium.Now when we have found out the speed of light we say that one metre is a given distance that light travels in one sec.
So with this we have two out of three , we have second as a unit of time , we have the total speed of light which light can travel in that one second and then we have a very small distance which light has covered in that one second while traveling with it's fixed speed so we now call that one metre.

Even though I believe that this accuracy for metre is more like a physics theoretical problem than a real one.It was just an academic hole that needed to be solved rather than a real problem.
It's like fixing all the currencies to gold so that we have one pretty much universal point of reference because without a accurate point of reference everybody can be right and wrong in the same time...

 

[technician]

'One metre can vary depending on the man doing the measurement'
'On the measuring tape'
'A physics theoretical problem'
'An academic hole that needed to be solved rather than a real problem'

This is a physics forum and metres, seconds, kg etc are defined in some way.

They are not just theoretical problems or academic holes.

We have to be using the same language here !

 

[jbriggs444]

Not really. They understood that the speed of light was a fundamental constant. So instead of having two units defined independently and with some independent uncertainty, the left just one - the time unit - defined with an uncertainty, and the other one related to it exactly via the constant.

http://en.wikipedia.org/wiki/History_of_the_metre

As I read it, this article takes the position that it is the accuracy of the methods for realizing a length measurement that have driven the redefinitions of the meter first from the artifact standard based on a hunk of metal to a natural standard based on the a krypton wavelength and then finally to the current standard ultimately based on a cesium frequency and the speed of light.

Paraphrasing [and hoping I'm not mis-reading] it was the accuracy of interferometry drove the first redefinition. It was the ability to trace interferometric measurements back to a time standard that drove the second.

 

[D H]

So...am I correct in saying that now the speed of light is a defined quantity and that the metre is a derived quantity?
Or...is 'derived' the wrong word to use in relation to the metre.
Can you recommend any references in this area?

The definitive source is the International Bureau of Weights and Measures (BIPM; the official acronym is French). Web site: http://www.bipm.org/en/home .

As far as 'derived' is concerned: Look to special relativity. One way of looking at the Lorentz transformation is that it is a rotation (a hyperbolic rotation) in Minkowski space. This in turn can be viewed from two different perspectives:
- It's a mathematical trick just that happens to work, or
- It's a true representation of the underlying physics.

The scientific consensus is that it's anything but a trick, that time and distance are indeed different aspects of the same thing. Fix one and the other *has* to be a derived quantity. We've chosen time because it is more readily measured. To be consistent with physics, distance (the meter) must necessarily be a derived quantity.

This begs the question, how many different independent dimensional quantities are there? Attacking distance is but the start of this unification process. Of the seven base units, there should be just two independent units. Luminosity is a measure of power specific to the human eye, temperature is a measure of energy density, mass is a measure of energy, current is a measure of electrons per second. All that's left are the second and the unitless mole, which is an arbitrary constant that relates the quantum world to the macroscopic world. This means that there's only one independent base unit, plus one arbitrary unitless constant.

The BIPM is on the path to making the second the sole independent dimensional quantity by making Planck's constant, the elementary charge, and the Boltzmann constant defined quantities. Whether this will happen anytime soon is a different question. Metrology is a rather conservative science.

 

[johnbbahm]

As to the second question, about how good of a vacuum do you need.
That falls to something called mean free path,
http://hyperphysics.phy-astr.gsu.edu/hbase/kinetic/menfre.html
As long as the mean free path is greater than the confines of your measuring
device, you are close enough.

 

[technician]

The definitive source is the International Bureau of Weights and Measures (BIPM; the official acronym is French). Web site: http://www.bipm.org/en/home .

As far as 'derived' is concerned: Look to special relativity. One way of looking at the Lorentz transformation is that it is a rotation (a hyperbolic rotation) in Minkowski space. This in turn can be viewed from two different perspectives:
- It's a mathematical trick just that happens to work, or
- It's a true representation of the underlying physics.

The scientific consensus is that it's anything but a trick, that time and distance are indeed different aspects of the same thing. Fix one and the other *has* to be a derived quantity. We've chosen time because it is more readily measured. To be consistent with physics, distance (the meter) must necessarily be a derived quantity.

This begs the question, how many different independent dimensional quantities are there? Attacking distance is but the start of this unification process. Of the seven base units, there should be just two independent units. Luminosity is a measure of power specific to the human eye, temperature is a measure of energy density, mass is a measure of energy, current is a measure of electrons per second. All that's left are the second and the unitless mole, which is an arbitrary constant that relates the quantum world to the macroscopic world. This means that there's only one independent base unit, plus one arbitrary unitless constant.

The BIPM is on the path to making the second the sole independent dimensional quantity by making Planck's constant, the elementary charge, and the Boltzmann constant defined quantities. Whether this will happen anytime soon is a different question. Metrology is a rather conservative science.

I know about the international bureau of measurements.
As far as derived units are concerned it is not necessary to resort to principles of relativity to clarify or illustrate this.
Area, volume, pressure etc etc are all derived, common physical units.

 

[Nugatory]

I know about the international bureau of measurements.
As far as derived units are concerned it is not necessary to resort to principles of relativity to clarify or illustrate this.
Area, volume, pressure etc etc are all derived, common physical units.

You may be missing the point of DH's post - he is saying that at least one of distance and time ought to be a derived unit, and that conclusion does require resort to principles of relativity.

 

[Crazymechanic]

Never thought that my interest in the speed of light would spark such a lively discussion.
Great.

As to technician , when I said the phrases "academic hole" and that the measuring of the metre is uncertain I meant that all of those where human derived quantities , nowhere in nature you have a natural some abstract phenomenon called 1 metre , it was us who made it up in order to somehow do the same measurement twice and with similar certainty as the first time.
Now realizing that the speed of light is a fundamental constant of nature was something different and I guess that why they related the metre figure to it.
And by an academic hole I meant just that.
It's always better if you have a somewhat absolute point of reference which you can then cut in smaller parts than if you have many many small ways of measuring something which in the end all stand only by themselves, I think this is the most important point , even though it was not my original question I enjoy this discussion.

 

[Dale]

So...am I correct in saying that now the speed of light is a defined quantity and that the metre is a derived quantity?
Or...is 'derived' the wrong word to use in relation to the metre.

I think that the words you are looking for are "exact" and "uncertain". The speed of light is defined to have an exact value, it is not measured. The measurement of a number of meters is subject to experimental uncertainty.

 

[Buckleymanor]

I think that the words you are looking for are "exact" and "uncertain". The speed of light is defined to have an exact value, it is not measured. The measurement of a number of meters is subject to experimental uncertainty.

It's a bit chicken and egg though.How can you define the speed of light to have an exact value without more accurate clocks and metre devices.
Even then there is allways room for improvement.

 

[BruceW]

the point is that we define the meter and the second such that the speed of light has an exact integer value. So in the SI units, we know the speed of light exactly, without having to do any measurements whatsoever.