I heard a few times that using them as sunglasses is hurting the eye since UV light is not filtered, but the pupil is wider than it would be w/o wearing them because the visible light is dimmed.

I have always assumed since atmospheric (i.e. not reflected from flat surfaces) sunlight is unpolarised and since the glasses used in 3D cinema only allow circular polarised light to pass they do indeed filter out a large portion of the intensity of the incoming (unpolarised) light.

Does this not work for the UV part of the spectrum or is my initial assumption wrong?


The use of anything but properly designed sunglasses is very foolish and poses great risks to your long term sight, and maybe for reasons that many people do not wholly appreciate.

First of all, let's write down what unpolarised light is. We choose two basis polarisation states: let's go with left and right polarised light in this case since you say that 3D goggles filter these (I actually don't know what 3D goggles do: since I am blind in one eye I can't see 3D, so I've never given it much thought because I have no intuitive grasp of what 3D sight "looks like"). A general polarisation state can be represented by a vector of two complex numbers $E_+$ and $E_-$ standing for the amplitudes of the left and right components of the light and unpolarised light is where they constants vary swiftly with time: not too fast relative to the light's frequency but still fast enough that over any time period long enough to get a possibly harmful dose of light in the eye. Therefore, for the purposes of this argument we can think of the average total intensity of the light as made up of the average separate intensities of the two polarisations and, for unpolarised light, the averages of the two basis state intensities are equal. The total light intensity is thus $|E_+|^2 + |E_-|^2$ and once you knock out one polarisation state contribution with your 3D glasses, you block half the light. This result works for ANY choice of basis state pairs, so the same conclusion arises from the use of linearly polarising sunglasses. I'm not sure what you're thinking of when you say "3D glasses filter out a large proportion", but I'm guessing that you're thinking it's much more than one half. It has to let something through, otherwise the glasses would be blank! So half the UV light is still getting through your 3D glasses and this is well enough to cause long term damage that can be prevented by proper sunglasses designed to filter UV. That's assuming that the 3D glasses still work as circular polarisation filters im UV, which, as you say, is not certain.

As you rightly point out, the safety benefits of sunglasses are wholly owing to their working as ultraviolet filters. There are two risks to the eye from intense light:

  1. The first is the thermal risk, where there is too much energy getting into the pupil and the local heating at the focal disk on the retina destroys the retinal tissue. Contrary to commonly held beliefs, the damage risk from this effect in normal sunlight conditions for a HEALTHY retina is minimal - EVEN if you like migraines enough to stare straight at the Sun continuously! (there are exceptions for retinal disease or dysfunction - so it's not a good idea to test this assertion out). With our pupils shrunken to their smallest size (about 1mm diamter) at noon on a clear day, staring at the Sun will deliver a heat dose of about 1 mW to the retina. This heat dose is WELL within the capabilities of a healthy retina to dump: the retina is superbly well envasculated and healthy blood flows carry off even this level of heat with almost no chance of any damage. Interestingly, laser safety standards for longer wavelength visible light set the intrinsically safe level at about the same level we experience when staring staight at the Sun. You will of course be swiftly temporarily blinded by such staring though as the ATP energy reserves of your cone and rod cells are drained owing to overstimulation, so we have a strong phsychological aversion to such staring - from an evolutionary standpoint it is not good for either predators or prey (we are both) to be blinded for minutes at a time! As an aside, the danger of looking at a total solar eclipse is that the pupil only responds to average light levels and swells to 50 times its area (7mm diameter) in the near-darkness of a Solar eclipse: you are therefore at significant thermal eye damage risk if look at the "diamond ring" and following bright stages of an eclipse after totality, when you can easily get thermally highly dangerous doses of 20mW or higher.
  2. Here's the main issue with sunglasses. Longer wavelength UV will cause sunburn of the retina at even low but constant levels. Shorter wavelength light begets phototoxicity: this arises when the light is energetic enough (photons of short enough wavelength) to knock outer shell electrons off and bring about chemical change and thus real cellular damage (even cellular nuclear damage, with its attendant cancer risk) within the retina. You wear sunglasses to avoid the painful longer UV wavelength sunburn that you'll inevitably get at chronic exposure to even low level sunlight and avoid the medium and long term risks of cataracts or worse (even eye cancer) that arise from chronic exposure to even low levels of shorter wavelength UV. The factor of one half attenuation I cited above for polarising filters does very little if any good in mitigating these risks.

So forget about sunglasses being useful for their general power attenuation effects. The power alone of sunlight in normal conditions (i.e. not an eclipse) is an almost nonexistent risk. It's all to do with phototoxicity and sunburn, so you need properly designed sunglasses that can attenuate the relevant UV bands by several orders of magnitude. You might get some relief from the discomfort of squinting in bright light, but as shown above, that discomfort is almost wholly psychological and as you point out the filters will likely raise the UV dose to your eyes owing to the swelling of pupils in low light levels. You can't feel the real damage proper sunglasses avoid till it's far too late.

  • $\begingroup$ I'm... not sure I understand this answer. I don't think the OP was suggesting looking directly at the Sun - that would be foolish with or without sunglasses. And certainly walking around outside with nothing on is safe, so walking around with 3D glasses on is no worse, if a bit weird. Besides, any glasses you get that cost less that $500 are UV-opaque, since they will either be made of non-quartz-crystal glass or of organic-double-bond-rich plastic. "UV-blocking" is just a marketing ploy to get people to pay more money. $\endgroup$
    – user10851
    Dec 8 '13 at 20:01
  • $\begingroup$ @ChrisWhite All I'm saying is that any risk from the sun is photochemical - glare and "amount of sunlight" is not a problem. I was making the point that looking straight at the sun, whilst highly unpleasant, is not dangerous in the short term for a healthy retina - I was making the point that it is not the intensity or the optical power (under normal conditions) that is the risk. A minute or so will give you a dose of UV-B that will cause painful sunburn and greatly increase longterm risk of cataracts but it certainly won't blind you ... $\endgroup$ Dec 8 '13 at 21:02
  • $\begingroup$ @ChrisWhite ... Everything you say is right in the short term - there certainly is a well established link between UV in the sunlight and conditions like cataracts and so forth. The risks are all later-in-life risks. I don't know about the US, but here and in Europe there are simple standards that anything sold as "sunglasses" must by law comply with - these include UV opaqueness. So the "marketing ploy" is neutralized: you can buy effective sunglasses for under $10. I would suggest that wearing 3D glasses for a day in the snow or at the beach in .... $\endgroup$ Dec 8 '13 at 21:08
  • $\begingroup$ @ChrisWhite .... low lattitudes at midsummer leaves a painful UV damage that takes days to disappear, for this is exactly what happens if you don't shield your eyes from UV in these conditions. $\endgroup$ Dec 8 '13 at 21:09

I heard a few times that using them as sunglasses is hurting the eye since UV light is not filtered, but the pupil is wider than it would be w/o wearing them because the visible light is dimmed.

IFAIK, there is no evidence for this claim. See this paper (unfortunately, it's behind a paywall):

The supposition that, because of pupil dilation, there are greater influxes of solar UV and short-wavelength visibile radiation when some sunglasses are worn is wrong. It is based on an incomplete and, therefore, misleading analysis. That analysis is particularly defective in ignoring the attenuation of the more highly actinic shorter wavelengths. This paper provides the most nearly complete analysis that is now possible. It treats both the geometric and radiometric aspects of ocular irradiation. A survey of over 400 retail sunglass lenses of all types showed that all would reduce UV radiation densities in all parts of the crystalline lens and on the retina. With spectral actinic weighting, the effective reductions would be even greater.

On the other hand, there has been a long discussion about sunglight dangers, but there is no strong evidence which proves any eye risks.


Well, I have a UV radiation meter used for children protection and so I tested the UV filtering capacity of all glasses I could find. All policarbonate lenses, even the cheapest are capable of blocking UV rays from level 12 to 0,2 or 0,3. The color has no impact in the filtering capacity, even white lenses are capable of blockinh UV rays by the same magnitude. When I tested the polaroid lenses of #D glasses I was also astonished by the similar filtering capability, maybe that's an effect of the plastic used, but the model I tested indeed blocks UV rays. Anyway, why would you use a 3D glasses for sun protection? They 're simple horrible and terrible for your nose, so please get a nice sunglasses, you can find very efficient cheap models.


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