Improving mismatched DIC parts with oblique illumination

Beatsy · Post by **Beatsy** » Fri May 02, 2014 9:01 am

In the interesting discussion about condenser NA < objective NA, oblique illumination was mentioned. That prompted me to take a fresh look as I tend to neglect it in favour of DIC. Bad practice for a wannabe diatomist

On my Zeiss, I use an NA 1.4 condenser front lens with the aperture wide open to get light from the offset condenser into high-power objectives. As discussed before, the 1.4 doesn't work well with most DIC slider/prism combinations for me - the DIC effect is faint (at best).

The scope was still configured for DIC when I first set up for oblique and I was amazed to see that a slider, prism and objective combo that normally produced a narrow dark bar in the background had reverted to a smooth gradient with a nice DIC effect (in addition to the oblique effect). So I tried other combinations and found many, though not all, showed improvement.

It's still hit and miss, but adding oblique to the mix gives me more usable DIC slider/prism/objective combinations than I had with straight Koehler illumination. A happy accidental discovery!

There are a couple of down-sides. Much reduced light and image shifting which distorts stacked images a little. But overall, I really like the images it produces.

Have I just rediscovered a well-known practice? Anyone else tried this? Why does it work?

Here are some comparison shots (25-image focus stacks). All were taken through a dry Plan-Neofluar 63/0.9 with the correction collar adjusted for the top surface of the specimen (nearest the cover slip).

Just brightfield

Just oblique illumination (got a bit of glare over the specimen - oops)

Just DIC (Koehler illumination) - it's quite low contrast with this condenser, slider and objective combo, but still usable by shooting RAW and adjusting levels in Photoshop.

Oblique + DIC. Oblique direction and DIC shear are both oriented along the length of the specimen. The ends got slightly elongated in Zerene due to pretty-extreme image shift as focus changes.

JohnyM · Post by **JohnyM** » Fri May 02, 2014 12:13 pm

Sure!
This method has two improvements. It add an obligue light effect via selective aperture shutting and also "cutting" the interference fringe. It's hard to explain with my english, but i'll try another way.
Setup your scope to DIC normally and then remove condenser prism and one of the eyepieces (you can insert auxilliary microscope [phase microscope] if you have one). You can usually see a one narrow black fringe and sometimes few colour on the peripheries. Now you can just make a slit diaphragm and put it under the condenser to "cut" the dark interference fringe and voila - you got an low resolution DIC (so called rediscovered "PlasDic" work like this).
Now remove the slit and put the condenser prism into lightpath. You can see that the previously narrow dark fringe is now much wider or (ideally) fully filling the aperture. If not, when you set an obligue light, you just cut it inperfections on the periphery.

Charles Krebs · Post by **Charles Krebs** » Fri May 02, 2014 12:15 pm

On my Zeiss, I use an NA 1.4 condenser front lens with the aperture wide open to get light from the offset condenser into high-power objectives.

Quick question.... are you getting "oblique" by de-centering the entire condenser or by de-centering the the condenser aperture?

Generally it is not the best idea to actually offset the condenser, but to off set the aperture.

A quite detailed discussion is found here:
http://micro.magnet.fsu.edu/primer/tech ... intro.html

Oblique is a very under-used illumination technique. It is very unfortunate that most condensers on current and recent models are held by their base, and have no access to a filter drawer or the plane where the aperture is located. In older microscopes most condensers were held by a collar near the top which allowed for a readily accessible filter "tray" that could easily be utilized for oblique and Rheinberg illumination.

I used obliques extensively before I had DIC and still do with brightfield and cross-polarized light. I have only played around briefly with it in combination with DIC, but did not find it to be that useful there... perhaps a more thoughtful look is warranted.

Some spectacular results using a what is basically oblique with some clever innovations have been posted by users Litonitus and Jacek. Litonitus dubbed his creation the "UFG" and you can read more about it here:

http://www.photomacrography.net/forum/v ... hp?t=15142

A little searching in the forum will provide some extraordinary examples of what can be accomplished.

It is also interesting to look at some other methods of "modified" oblique, such as the "Mathias Arrow"

Http://www.schwaben.de/home/mathias/english.html

http://www.microscopy-uk.org.uk/mag/ind ... lique.html

Ichthyophthirius · Post by **Ichthyophthirius** » Fri May 02, 2014 1:04 pm

Hi,

This is interesting. To understand the observations, it's worth looking back at one of the early interference contrast configurations.

In this configuration, a Wollaston prism is situated in the back focal plane of the objective. The interference fringes can be seen when looking through the tube of the microscope (Fig. 7.7c; see below). When looking through the eyepiece, you would see an indistinct black bar running through the image.

To generate differential interference contrast, the higher order fringes have to be blocked out, so only the central fringe (S') remains illuminated. This is done by introducing a slit diaphragm (Dc) in the front focal plane of the condenser. The diaphragm has quite a narrow central slit (Wc) in the same direction as the objective Wollaston prism fringes.

This configuration will give DIC! This is interesting to know; only the objective side prism is strictly necessary. PZO (Poland) manufactured DIC sets that had a special condenser with variable slit aperture. Zeiss currently offers the PlasDIC system which also uses a condenser slit diaphragm.

The disadvantage of this set-up is that resolution perpendicular to the direction of the slit is severely reduced. Also, out-of-focus objects appear no longer round but rod-shaped which looks bad in photography.

This is why in the Nomarski DIC set-up, a Wollaston prism is inserted instead of the slit aperture. This prism is simply a compensator for the fringes generated by the objective-side prism. This set-up allows the use of the full condenser aperture for illumination. If the compensation is correct, the entire back focal plane of the objective can be set to black.

Now, if your condenser prism doesn't match the objective prism, the higher-order fringes are not fully compensated and the back-focal plane cannot be set to black. The contrast is poor and the image looks washed-out.

When you set your condenser to oblique illumination, you effectively obscure the higher-order fringes and the contrast becomes stronger. Effectively, you have a mixture of compensation: part slit aperture, part compensating prism.

You should be able to get the same effect by inserting a home-made slit aperture, in the correct orientation, into the filter holder of your microscope, as close to the condenser as possible.

Regards, Ichthy

JohnyM · Post by **JohnyM** » Fri May 02, 2014 1:48 pm

That's, exactly what i wrote

.
It's also worth to bear in mind, that using slit you can make DIC with any prism and any objective and the quality depends on the slit dimension.
Only extremely narrow slit diaphragm give significant coherent noise (?) and it's lowering resolution mostly in the direction which wouldn't be resolved anyway (due to DIC resolution directivity), look at the plas-dic zeiss article.
Slit was also commonly used with mach-zender interferometers (like interphako).

Beatsy · Post by **Beatsy** » Fri May 02, 2014 1:57 pm

Charles and Ichthyophthirius - regarding "filter trays" and offsetting condenser aperture rather than condenser - my setup doesn't properly support those things.

I can offset the aperture, but only a little (enough to center phase rings or align the iris with the objective). However, I can move the whole condenser a lot more so this is what I do for my "oblique". I know it's not ideal as it gets me out of line with the filament and the best part of the condenser lens, but it seems to work OK. I tried partially rotating the phase/dic/etc selector to push the aperture to one side, but being a circular hole, it doesn't light enough of the aperture when shifted sideways so I lose resolution and don't get much of an oblique effect anyway.

I read about the UFG with great interest, but it wouldn't work in my filter slot as this is not aligned with any of the light source conjugate planes. Any stops placed at that position appear as fuzzy shadows in the image plane. I think the only place I could put homemade stops is in the spare (number 3) slot inside the condenser itself. A project on the list...

Here's a pic to explain the bits I'm referring to...

Beatsy · Post by **Beatsy** » Fri May 02, 2014 2:03 pm

PS. Johny - thanks for the info on PlasDIC. I'd read about it before, but not how it works. Helps putting the puzzle pieces together.

Almost embarassed - had this scope ten years and STILL learning stuff about it. That's what makes it interesting though

Ichthyophthirius · Post by **Ichthyophthirius** » Fri May 02, 2014 2:23 pm

JohnyM wrote:That's, exactly what i wrote

I know. You type much faster than I do