epi cross polarization purple highlights

A forum to ask questions, post setups, and generally discuss anything having to do with photomacrography and photomicroscopy.

Moderators: rjlittlefield, ChrisR, Chris S., Pau

dlg
Posts: 26
Joined: Wed Nov 04, 2015 11:47 pm

epi cross polarization purple highlights

Post by dlg »

I've just been fiddling with using cross-polarization to cut specular highlights from EtOH-wet preserved flowers under a Zeiss Discovery V20 with planapo 1.5x. Flowers are not submerged, but EtOH wet, so a worst-case for highlights. Analyzer on lens is from Zeiss, polarizers on Zeiss LED fibre optics are from Schneider (I think they provide them to Zeiss anyway). At maximum extinction, I get fairly distinctly purple highlights, at reduced extinction, they go yellow.
With less extreme examples (fresh flower with waxy cuticula) I do not notice any color artifacts. I have not noticed it on dry polished metal.

- Is this just unavoidable, given that no polarizer is perfectly neutral grey?
- Is this a problem due to unmatched polarizer and analyzer?
- Could this be improved by higher quality (which?) polarizing filters?
- Could that be due to LED fibre optics source?

I lean towards first option, but just wanted to check in re other experiences.

g4lab
Posts: 1437
Joined: Fri May 23, 2008 11:07 am

Post by g4lab »

- Is this just unavoidable, given that no polarizer is perfectly neutral grey?
Prism Polarizers such as Nichols or Glan prisms are colorless and were what got supplied with pol scopes back in the day.
- Is this a problem due to unmatched polarizer and analyzer?
You are observing birefringence or retardation in the specimen itself. The fact that they go from magenta to yellow is pretty much proof of that fact.
- Could this be improved by higher quality (which?) polarizing filters?
The phenomenon proves the good quality of your polarizers.
- Could that be due to LED fibre optics source?
Not likely but you should check your light source for color shifts when you rotate the polarizers. I have not yet seen anything on the polarizing state of LED emissions. Maybe someone else on the list has run across something. But rotate you polarizers and analyzer both crossed and uncrossed without specimens and see whether it causes any color drift.
I have not noticed it on dry polished metal.
Metal is electrically conductive and does not preserve polarization. Pol effects are only seen on insulators. Like the surface of water or alcohol.

rjlittlefield
Site Admin
Posts: 23564
Joined: Tue Aug 01, 2006 8:34 am
Location: Richland, Washington State, USA
Contact:

Post by rjlittlefield »

g4lab wrote:Metal is electrically conductive and does not preserve polarization. Pol effects are only seen on insulators. Like the surface of water or alcohol.
Strictly speaking the reason for the observation is the other way around: most metal preserves polarization for all reflections, where insulators preserve it only for specular reflections from the surface, and not for subsurface "diffuse" reflections from the body of the material. This is why cross-polarization can kill the surface reflections while still letting through the others. Crossing to extinction with most polished metal subjects will make the whole thing go dark. (See http://www.metallography.com/technotes/ ... aging2.htm for a discussion of some exceptions due to non-cubic crystalline atomic structures.)
dlg wrote:- Is this a problem due to unmatched polarizer and analyzer?
- Could this be improved by higher quality (which?) polarizing filters?
From what you've described, I consider both of these to still be possibilities. The shift from purple to yellow could simply be a matter that the illumination is naturally yellowish while the polarizers go purple near extinction. A good check is to make a simplified system in which you look directly at your light source through the crossed polarizers. If the color still goes purple near extinction, regardless of the orientation of the light source, then the problem is in the polarizers. Higher quality polarizers that do not have this problem are available from sources such as Edmund Optics, and undoubtedly others.

--Rik

dlg
Posts: 26
Joined: Wed Nov 04, 2015 11:47 pm

Post by dlg »

Thanks for your replies. Promoted some reading on prism polarizers. Interesting, but possibly a bit overkill on my application.

The LED light source is remarkably color stable, and intensity adjustments have close to zero effect on color temperature, I checked that with a Minolta colormeter III. That is one of the nice advantages of LED over incandescent bulbs.

I did try rotating the analyzer then re-adjusting polarizers, and it did not result in any change in the color patterns observed.

So the easiest will be to view the specimens fully immersed in liquid.

Pau
Site Admin
Posts: 6053
Joined: Wed Jan 20, 2010 8:57 am
Location: Valencia, Spain

Post by Pau »

Could you post a picture illustrating the issue?

I'm thinking in two other possibilities:
- despite using a planapo with a high end Zeiss Discovery stereo it could be chromatic aberration due to the optics or to the refraction in the liquid on the subject.

- In many LEDs, specially those with a matrix of actual LEDs covered with a common yellowish phosphor the light emitted by the phosphor between LEDs is yellowish and the light emitted by the phosphor just over the LEDs more bluish. I often observe it when I use one of these LEDs as light source in my microscope. A diffuser disc just before the illuminating polarizer will average it and kill the effect, but I think that fiber optics could do the same.
Pau

Lou Jost
Posts: 5948
Joined: Fri Sep 04, 2015 7:03 am
Location: Ecuador
Contact:

Post by Lou Jost »

In my experience, most photographic polarizing filters are not not 100% efficient at short wavelengths. A good way to fix that is to eliminate the short wavelengths in your light source. The "fringe-killer" astronomical filter discussed in a recent thread might fix this completely if that is the problem. A short-wavelength cutoff filter would also eliminate this.

dlg
Posts: 26
Joined: Wed Nov 04, 2015 11:47 pm

Post by dlg »

No pictures yet, will follow up with that in the next couple of days.
I don't think it is chromatic aberration. The purple/yellow is not a fringe (seen that Leica scopes), but a solid area.

Re short wave cut-off, I would think that regular (non quartz) glass in filters and fibre optics should cut off UV.

The effect of the liquid droplets could very well have to do with it. In that case, no optics will be able to solve it.

The polarizers are mounted onto focusing lenses on the tip of the goose necks. I de-focus it, so that should have the same effect as a diffusor disc.

Lou Jost
Posts: 5948
Joined: Fri Sep 04, 2015 7:03 am
Location: Ecuador
Contact:

Post by Lou Jost »

dlg, I was referring to visible violet, not UV. Get rid of the violet light and you should also get rid of the violet highlights.

rjlittlefield
Site Admin
Posts: 23564
Joined: Tue Aug 01, 2006 8:34 am
Location: Richland, Washington State, USA
Contact:

Post by rjlittlefield »

rjlittlefield wrote:A good check is to make a simplified system in which you look directly at your light source through the crossed polarizers. If the color still goes purple near extinction, regardless of the orientation of the light source, then the problem is in the polarizers.
dlg wrote:The LED light source is remarkably color stable, and intensity adjustments have close to zero effect on color temperature.
The issue that I was raising doesn't have anything to do with changing intensity adjustment of the light source. When I wrote "near extinction", I was talking about having the polarizers at or very near 90 degrees, perfectly crossed. What often happens under those conditions is that relatively more purple/blue light gets through the crossed filters, leading to a change in the light color. It's the same effect that Lou Jost is talking about.

Since your filters are mounted on the ends of your fiber bundles, this effect can be easily tested by simply looking directly at the filtered fiber through another filter, and rotating the viewing filter until the fiber ends go as dark as possible. I'd place a modest bet that when they go dark, they also go purple/blue. Do not shine the fiber onto or through anything else, like a piece of white paper or a diffuser disk, since that will destroy the polarization. The optics chain should be just fiber, crossed polarizers, eye.

--Rik

Lou Jost
Posts: 5948
Joined: Fri Sep 04, 2015 7:03 am
Location: Ecuador
Contact:

Post by Lou Jost »

What often happens under those conditions is that relatively more purple/blue light gets through the crossed filters, leading to a change in the light color. It's the same effect that Lou Jost is talking about.
If you do the test that Rik suggested, and if it turns out that blue and purple leak through the crossed polarizers, you can improve things by cutting out the blue and purple light of your light source using a filter. Since your flowers are in alcohol I would think the color doesn't matter to you anyway. There will probably still be some trace of the highlights but they will be dimmer relative to the non-highlights.

I should add that I always photograph alcohol-preserved flower while still submerged, to preserve their shape and also to eliminate reflections. Here is an example of an orchid's anther cap (it belongs to a new species of Maxillaria I am describing), submerged in alcohol; photo with Mitu 10x/0.28 on 135mm Vivitar (Komine) tube lens. Background is glass with Protostar behind it. Lit with two well-diffused Yongnuo flashes placed laterally and slightly below photographic plane, to minimize background and subject reflections. Please pardon some unretouched dirty-pixel streaks. This is a quarter of the full image.
Image

Post Reply Previous topicNext topic