Does the answer change with the angle of incidence? If specular reflectance changes with angle of incidence, what is the relationship between them? If such surfaces are possible, where can I find a source?

  • $\begingroup$ Total internal reflection only works at large incidence angles. Not sure the specific numbers. $\endgroup$
    – user253751
    Jul 27 at 17:02
  • 4
    $\begingroup$ Something like thorlabs.com/newgrouppage9.cfm?objectgroup_id=12944 (>98.5% reflectivity over 250-272nm at 45 degrees)? $\endgroup$
    – Jon Custer
    Jul 27 at 17:19
  • $\begingroup$ Polished aluminium has a high reflectance down to about 200nm. In my PhD days we used to vacuum deposit aluminium on silica flats to make mirrors for uv, though I don't think we went down as far as 250nm. But given how cheap and easy it is I suggest it's worth a try. $\endgroup$ Jul 27 at 17:49
  • $\begingroup$ here for example: layertec.de/en/shop/articles-by-coating_95a3fa9888d5139d478f/… $\endgroup$ Jul 27 at 18:04
  • 1
    $\begingroup$ @JohnRennie the link I posted is for a dielectric coating rated for 99.5%. $\endgroup$ Jul 27 at 19:09

1 Answer 1


Your question seems to have two parts.

Can you get the reflectivity > 99% The answer is yes. Dielectric mirror stacks can give you that.

Does the reflectivity depend on angle of incidence The answer is also yes, and also you may want to consider polarization.

As an example this is a mirror designed for the 4th harmonic of a Yb laser. It is a dielectric mirror with high reflectivity and it is designed for the mirror to be placed at a 45 degree angle with respect to the incoming beam. for your wavelength both the S and P are probably greater than 50%. It is about 200 dollars, and probably not that large probably 1 inch in diameter.

If you don't have this mirror at 45 degrees then it is a little more complicated, you can order a mirror for the angle you want, or you can consider that as you change the angle of the mirror the effective length in the dielectric stack is changing the path length in the stack. This effectively shifts the reflectance spectrum with the angle. Something like 1/cos($\theta$). So for small angles it doesn't shift much.

Thor labs UV mirror

If you want to have >99% reflection from a wide range of angles the it more difficult. If the angles start to large you will be outside the bandpass of the dielectric mirror.

Pure Aluminum won't get you the 99% but is probably the best compromise for something that can be large area to get a specular reflection and it can shaped, for example by diamond turning to get a specular reflection. And as pointed out in the comments it can be evaporated. It will have a pretty flat curve for reflectivity as a function of angle.

However Aluminum will oxidize over time and lose its reflectivity some and have increased scattering. Most optics companies will recommend a protective coating and those may be a few layers of dielectrics. This brings you back down to the 90 ish percent range.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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