3
$\begingroup$

I am getting very confused about the reason why dichroic mirrors should be placed observing a certain orientation. I read everywhere that you should place the mirror with the coated side facing the incident light (fair enough), but could not find any clear explanation why. What happens if you orient the mirror in the 'wrong' direction? Could maybe someone point me to a relevant resource?

Note that I don't have any means at the moment to do the experiment and see by myself.

$\endgroup$
1

3 Answers 3

2
$\begingroup$

If you flip the optic around, you’ll get a ghost reflection from the first (uncoated) surface and then possibly additional effects from multiple reflections within the glass substrate. Nothing will happen to the transmitted wave (transmission is reciprocal).

$\endgroup$
1
  • 2
    $\begingroup$ If it is a high power mirror, hitting the back (uncoated) surface may result in bad things happening to the substrate... $\endgroup$
    – Jon Custer
    Jan 1, 2022 at 2:53
0
$\begingroup$

Dichroic mirrors work using several layers of thin-film coatings that cause select wavelengths to have constructive interference at the reflected angle and destructive interference when transmitted. The reflection angle is almost always designed to be 45 degrees. The wavelength at which complete constructive interference changes as the angle of reflection changes.

Put simply, changing the angle of the dichroic changes the wavelengths that are reflected or transmitted.

Some dichroic systems actually leverage this angle effect to create a continuously variable filter for hyperspectral imaging.

$\endgroup$
0
$\begingroup$

It's complicated. You must do all the math and logic yourself.

Total internal reflection might occur, given certain angles of certain rays.

What is the index of refraction of the ambient medium (air, or water, or vacuum, usually)? The substrate? Get your "Snell" on, or Harriot or Descartes, or Ptolemy, if you prefer.

The previous answers are a bit better than mine. These experts will know what the HIR and LIR alternating layers are doing, specifically.

I am pretty certain polarization will matter. But if you have a plane wave, monochromatic, single phase, linearly polarized, at normal incidence (single-mode), maybe this won't matter much, or at all.

But, filter backwards, almost certainly, 4% reflection at every air-to-glass interface (more, quickly, steeply, with increasing angle of incidence), and see the classic Born and Wolf treatise on physical optics, to see differential reflection of the TM and TE portions of the incident rays. With incident angle, they are increasingly reflected.

I am a mere biologist, and so, uncertain about the "coupling" effect between TM and TE E-M fields. I assume, if one is knocked down 5% and the other 3%; the whole thing is knocked down 5%? Cuz? They are mutually-coupled-and-mutually-propagating? Again. I know very little, for how much I know.

So, backwards-filter equals "bad". And even though I suspect that the two answers above me are written by people with 10 X my knowledge and experience, I have trouble truly believing that "transmission is reciprocal" entirely, as the author appears to meaningfully proffer. Because at least 4% is reflected, when filter-backwards. So, 100% - 4% ; and then (say the filter is designed for : 50% blue-reflected and 50% green-transmitted); that will be only 48% Refl. and 48% Trans. . Whereas, hitting the coated surface, first, will guarantee 50% Refl. and 50% Trans., and of the 50% transmitted, some negligible amount will reflect traveling from glass to air (2nd surface) (assuming normal incidence of the transmitted, exitant rays, at the second surface; out of the substrate material and back into the surrounding medium).

An open question? I always heard that unpolarized light was an electromagnetic wave, vibrating in every possible plane. And to further complicate matters, every electric field oscillation is coupled to a magnetic field oscillation, traveling perpendicular to it, along the line or ray of propagation. So? Is light a bunch of concentric "footballs" of E field (rotated sine wave; rotated about the line of propagation), coupled to a bunch of concentric "footballs" of M field? I think So? I think we have to conclude this? But what about phase? Uh oh? it gets more complicated. I'll stick to biology, than you very much.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.