Resolution improvement at same NA - achromat vs apochromat

[josmann]

Hey folks,

I've been doing some comparison investigation between achromatic and apochromatic lenses and noticed something interesting. I've been going back and forth between a 10x DPlan (NA .25) and a 10x SPlanApo (NA .40) looking at the remains of a bit of algae. The algae cell structure has some features which are nicely spaced such that they are just resolved by the Apo lens but not quite by the Achro lens.

I decided that, to be more fair, I should set my condenser to .25 NA for both so that I can just compare the chromatic differences between the two. I was a little surprised, however, to see that I can still note a pretty significant resolution improvement with the Apo lens (not as good as setting the condenser wider, but still better).

So my question is what is the nature of this resolution improvement? My two guesses are:

A. The Achromat when "in focus" is actually slightly blurred since green and red/blue are both slightly out of focus. The Apo when "in focus" is actually in focus across most of the spectrum, so its able to resolve more. If I went to monochromatic illumination, I would see no major difference.

B. The sample itself is scattering light into higher angles such that, locally, the light coming from it has a higher NA than the condenser setting and the Apo is just behaving normally.

Appreciate any insights!

[MichaelG.]

Your answer (A) seems a good explanation to me … and you already know how to check.

Get those monochromatic sources ready !

MichaelG.

[viktor j nilsson]

I actually think that both explanations are in effect. Surely chromatic error will blur detail (A). But it is also true that some rays of light that have been refracted by the subject will hit the objective at a steeper angle (higher NA) than the light coming straight from the condenser. With the 0.25 objective, this light will be lost, but with the 0.4, the light will be picked up and contribute to image formation. This effect will be stronger for highly refractive objects. In macro photography, it is very clear that the effective NA and resolution depends on how much light is diffracted at steep angles by the subject.

Which of these effects are strongest in practice I don't know.

[josmann]

Your answer (A) seems a good explanation to me … and you already know how to check.

Get those monochromatic sources ready !

MichaelG.

Well I don't have a monochromatic source as such, but I do have a whole bunch of photo gels.

I stacked up 6 green gels and observed some RGB lights (high tech computer parts) and found that I was still getting a fair amount of blue transmission. I added two yellow gels to the stack after which I couldn't see much blue transmission and the green had that rather "532" shade to it.

And the results after filtering the lamp light with this were.... pretty much identical. It's maybe arguable that the DPlan improved a little bit, but the Apo was still clearly better resolved.

Now that I think about it more, I might be letting ray optics poison my brain a bit. I don't know what the physical length between the green and red/blue focal points is, but if they're all within each other's Rayleigh ranges (which would make a lot of sense from a design perspective), then the spot size difference for the same NA might be pretty negligible between it and the Apo. You would only start to notice the difference when things go out of focus and the difference is more exaggerated (purple/green fringing). Or maybe I'm just literally describing an apochromatic lens at that point - hmm... It's also possible that focal point differences like this might be more pronounced at higher magnifications - I don't have a 40x/100x Apo just yet, but I will soon...

I think I'm still struggling with the nuances of the condenser NA versus the diffraction/scatter that will naturally occur at the sample...

[josmann]

This effect will be stronger for highly refractive objects. In macro photography, it is very clear that the effective NA and resolution depends on how much light is diffracted at steep angles by the subject.

Yeah I think that has a lot to do with what's going on here. I'm actually not sure if this is a silica or cellulose I'm looking at but the resolved features are all totally refractive in nature. Thanks for the input.

[smollerthings]

I don't have a 40x/100x Apo just yet, but I will soon...

Check the working distance on these. My 40x apo was squishing my samples to death, generating fluid flows and disturbances to observation in a pond context. My plan fluor is much better suited.

[apochronaut]

Spherical aberration is a factor as well as in some cases, better contrast. While superior contrast itself does not contribute directly to resolution, poor contrast or less contrast might be a better term, can obscure resolvable details.

[MichaelG.]

I actually think that both explanations are in effect …

I’m sure you are correct Viktor … I just [wrongly it appears] assumed that checking the performance with monochromatic lighting would be the easy first step in the investigation.

MichaelG.

[viktor j nilsson]

I actually think that both explanations are in effect …

I’m sure you are correct Viktor … I just [wrongly it appears] assumed that checking the performance with monochromatic lighting would be the easy first step in the investigation.

MichaelG.

I think you are totally correct too - using monochromatic light seems like a good way to start to disentangle the effects of A and B from each other.

[75RR]

I'm actually not sure if this is a silica or cellulose I'm looking at but the resolved features are all totally refractive in nature.

You might find this article by David Walker interesting: https://www.microscopy-uk.org.uk/mag/ar ... -test.html

[josmann]

You might find this article by David Walker interesting: https://www.microscopy-uk.org.uk/mag/ar ... -test.html

Looks very useful, thanks! I actually do have one of those diatom test slides in the mail so this will be helpful when I start using mine

Check the working distance on these. My 40x apo was squishing my samples to death, generating fluid flows and disturbances to observation in a pond context. My plan fluor is much better suited.

Thanks, I checked out the literature on these just now (http://www.alanwood.net/downloads/olymp ... ctives.pdf). There are tables for each objective near the end of the document. The SPlanApo 40x and 100x have WDs of .13 and .15 respectively, so that's definitely not much margin! I'll have to be careful and survey the slides for any larger bits or critters before loading it in. Perhaps drop the stage before rotating the turret as well. The 10x and 20x by comparison are .55 and the SPlan 40x (which I've also got) is .5 which is pretty close but very manageable. I don't think Olympus made a PlanFL in a 40x variety, but they did make a LWD 40x which I might consider for use with deep well plates. You do pay an NA penalty, but the WD could be very nice. Overall, though, the apos produce just such a nice image in comparison to the achromats that I think I'll want to find a way to use them. I'm particularly fond of the lack of green/purple fringing for out of focus subjects which I've found pretty distracting on the achromats.

One intriguing objective for me is the 50x DPlan with a NA of .90 and a .23 WD. Dunno if anyone has experience with that but seems like a good middleground objective if you don't mind oil prep.

Spherical aberration is a factor as well as in some cases, better contrast. While superior contrast itself does not contribute directly to resolution, poor contrast or less contrast might be a better term, can obscure resolvable details.

Yeah I should mention that in both cases, my subject was restricted pretty much to the dead center FOV. In fact, changing from the 10x Dplan to my cheap 10x Chinese achromat (same NA) it's virtually impossible to discern any difference between the two - which was a little surprising in and of itself. By the way, I know you fancy yourself a bit of an optics whiz - above I speculated that the green focus and the red/blue focus for achromats might be designed such that they're within each other's Rayleigh ranges. Do you think there's any merit to that or am I way off base? Thanks!

[patta]

A third option
Not only chromatic aberration; the Apochromat was designed to work at NA 0.40, hence capable of higher resolution, thus needed a more sophisticate construction, which leads to smaller blur from aberrations and construction defects than the one with NA 0.25.
Instead for a standard 10x 0.25, the manufacturer won't bother to make it more accurate/corrected than the Raileigh limit for NA 0.25.

With this reasoning, also a 20x 0.40 (non-apo), with condenser at NA 0.25, should resolve a bit more than the 10x 0.25, since has smaller objective-induced blur.

Even more - aberrations, like Spherical, increase significantly with the NA. Again, the Apo was designed to have good performance at NA 0.40 (very hard job); and has an easier job at only 0.25, with spherical aberr further reduced.
Like, an expensive photographic lens f/1.4, stopped down at f/2.8, is sharper than a cheap lens at the same f/#.