Optimizing Tubular Optical Filter for Selective Photon Paths - Jim Adrian

[jamesadrian]

A tube having an inside diameter of 2 microns is one centimeter in length. A photon entering the tube will reach the sensor on other end only if its wavelength is not greater than 2 microns and the direction of its travel is does not cause it to be absorbed by the inner walls of the tube.

I wish to encourage any off-angle photons to bend away from the center line and become absorbed by the inner walls of the tube. I wish the angles reaching the other end to be as selective as possible, favoring only the photon paths that are most parallel to the tube. One idea is for the tube to have glass spheres in it that are 2 microns in diameter. This might bend off-angle photons toward the walls.

Suggestions are very welcome.

Thank you for your help.

Jim Adrian

 

[Drakkith]

What is the overall purpose of this device? What wavelength range do you want to capture? Everything under 2 microns, or only down to a certain wavelength?

 

[Baluncore]

If the walls are conductive then you have a circular waveguide. The cut-off frequency will not be exactly when wavelength equals the diameter. It will be dependent on the mode of propagation in the waveguide.

The length of your tube is unimportant. Any photon entering at one end will be guided by the wall. The directional sensitivity will be a characteristic of the entry structure.

It might be better to consider a solid optic fibre than an empty? tube. Then you can use available fibre coupling science and technology to design your directional filter.

An aperture that is one wavelength wide will have a beam width of one radian = 57°. You will probably need an interferometer array of apertures to get better resolution.

 

[Drakkith]

Does 2 micron wavelength radiation even obey the normal rules for waveguides?

 

[Baluncore]

Does 2 micron wavelength radiation even obey the normal rules for waveguides?

Where in Maxwell's Equations does scale get a mention ?

 

[Drakkith]

Where in Maxwell's Equations does scale get a mention ?

No idea. I was under the assumptions the frequency was too high at that range, but I must be thinking of something else.

 

[Baluncore]

It comes down to the particle : wave duality.
When the tube is a similar size to one wavelength is is a wave guide situation.
A larger diameter tubular collimator works with smaller particles wavelengths.
http://en.wikipedia.org/wiki/Collimator

 

[sophiecentaur]

It comes down to the particle : wave duality.
When the tube is a similar size to one wavelength is is a wave guide situation.
A larger diameter tubular collimator works with smaller particles wavelengths.
http://en.wikipedia.org/wiki/Collimator

Why try to invoke particles into this, just because the wavelength is small wrt the aperture? Wave optics works perfectly well for describing and predicting how light will behave in a Telescope with a metre diameter. How would you set about doing that with 'particles'?

 

[AlephZero]

I'm not too interested in arguments about the relevance of wave-particle duality, but I'm intrigued by how the OP plans to make a 2 micron diameter hole through the material (presumably "perfectly" straight) and then fill it with 2 micron diameter glass beads...

 

[Baluncore]

Why try to invoke particles into this, just because the wavelength is small wrt the aperture? Wave optics works perfectly well for describing and predicting how light will behave in a Telescope with a metre diameter. How would you set about doing that with 'particles'?

The wave particle duality requires that the photon be treated as a wave or a particle depending on the situation. The OP invoked the use of a very thin tube as a particle collimator. But a photon entering a thin tube must be treated as a wave in a wave guide and not as a particle. My wiki reference was to particle collimators based on very large tube diameters and lengths.

The functionality of a one metre diameter telescope is not due to the size of the tube, but to the mirrors and/or lenses supported by the 1 metre diameter tubular structure. If you remove the optical components and paint the inside of the tube black, then you have a crude particle collimator. The only “wave” effects of a large open tube will be diffraction at the edges of the entry and exit.

 

[sophiecentaur]

The wave particle duality requires that the photon be treated as a wave or a particle depending on the situation. The OP invoked the use of a very thin tube as a particle collimator. But a photon entering a thin tube must be treated as a wave in a wave guide and not as a particle. My wiki reference was to particle collimators based on very large tube diameters and lengths.

The functionality of a one metre diameter telescope is not due to the size of the tube, but to the mirrors and/or lenses supported by the 1 metre diameter tubular structure. If you remove the optical components and paint the inside of the tube black, then you have a crude particle collimator. The only “wave” effects of a large open tube will be diffraction at the edges of the entry and exit.

I think this is a false dichotomy. In all cases, diffraction is relevant - it's just the size of aperture that counts. Mirrors and lenses have edges and it is their finite size that determines the optics (which is usually contained in a tube, if you want good performance). This is just the same for a very small tube.

The 2d/λ factor will be what counts in the end and gives an indication of the diffraction limit or resolution you can obtain. It may not be the limiting factor for all set ups.

 

[Baluncore]

The 2d/λ factor will be what counts in the end and gives an indication of the diffraction limit or resolution you can obtain.

So you agree with me.

 

[sophiecentaur]

So you agree with me.

I guess I could do but wasn't sure what your post was getting at. I usually reckon that waves are the way to go except when QM insists otherwise. I.e. particles explain very few classical em phenomena well.

 

[sophiecentaur]

So you agree with me.

Btw, I don't agree with your distinction between a tube and a mirror / lens. The truncation of the wave front is precisely the same - the only difference is the introduction of a phase 'tailoring' by the optics.

 

[Baluncore]

sophiecentaur. Have you considered replying to the original post ?
Can you restrict your reply to the case specified of a 2 um wavelength entering a 2 um tube ?
Will the length of the tube being 1 cm long have any effect on the directional characteristics ?
Why ?

 

[sophiecentaur]

sophiecentaur. Have you considered replying to the original post ?
Can you restrict your reply to the case specified of a 2 um wavelength entering a 2 um tube ?
Will the length of the tube being 1 cm long have any effect on the directional characteristics ?
Why ?

Don't be a spoilsport, BC. As the OP never visited the thread again, I rather thought that the thread had license to go whither it wanted.

I don't think I have been very much off subject (no more than your average PF thread). I could not even start to approach the problem in terms of particles and rely on getting a valid answer. It's certainly not like firing peas through a pipe, which seems to be the 'ballistic' approach in the OP.
There have already been contributions which suggested a waveguide approach and I see no reason why it wouldn't work. A 1cm λ wave in a 1cm diameter tube should propagate. How it is intended to launch it is not known and I can't imagine that there is any loading inside the guide that could give a tight exit beam. the field across the exit of the tube would determine the beam width and I can't see how that could be improved upon. There is a lot of design information on circular waveguide feeds which would scale directly.

The fact that the OP seems to have left us makes me think that the question was probably a casual one and that there hadn't been a lot of previous research on the exact problem.

 

[Claude Bile]

The OP is essentially asking how to decrease the numerical aperture of the 2 micron waveguide.

Answer is simple. Put a lens in front of it, so rays that are marginally off axis become further deflected away from the optic axis.

Claude.

 

[sophiecentaur]

The OP is essentially asking how to decrease the numerical aperture of the 2 micron waveguide.

Answer is simple. Put a lens in front of it, so rays that are marginally off axis become further deflected away from the optic axis.

Claude.

This is all very well but how can you eliminate the effect of diffraction due to the aperture at the end of the pipe?
Also, how can you ensure that a 'useful' amount of power will actually enter at the source end? We're into the realms of feeding a mono mode optic fibre.