I always thought high RI mountants do not improve resolution but only increase contrast. The greater the difference between mountant RI and specimen RI, the stronger the contrast. Small, already-resolved features (on diatoms) become easier to see, but resolution stays constant.
Last week, while reading an old thread, I saw a comment stating that high RI mountants reduce illumination wavelength and improve resolution. Many images in the thread appeared to prove it too! So I thought I was wrong all along and this was my "new thing learnt" for that day.
Something still didn't seem right though and I couldn't stop thinking about it. As far as I knew, the formulae defining the "minimum resolvable distance" (resolution) have no term for RI. The only factors are numerical aperture (of condenser and objective) and the wavelength of the illumination.
So I did more searching and reading and now I think I've got it - at last!
The formulae mentioned above calculate lateral resolution, in x and y. We may say "resolution" in casual discussion but usually mean "lateral resolution" (whether we know it or not). In that case, what I believed remains true - high RI mountants do not increase (lateral) resolution, only contrast. But the formulae for calculating axial resolution (in z) do include RI as a term and axial resolution is increased by higher refractive index mountants. It all makes sense now. Except for one thing...
I can't explain why improved axial resolution looks like improved lateral resolution as it did in the images in that old thread - here: https://www.mikroskopie-forum.de/index.php?topic=40111 (https://www.mikroskopie-forum.de/index.php?topic=40111)
My working theory (wild guess) is that it only shows a visible improvement in resolution if the features resolved are very, very thin in the axial (z) direction. Like dots and slots inside the pores of Coscinodiscus sp. etc.
I am very interested to hear what others think or know about this. Thanks in advance.
Zitat von: Beatsy in Februar 26, 2023, 23:12:27 NACHMITTAGS
Last week, while reading an old thread, I saw a comment stating that high RI mountants reduce illumination wavelength
Hi Steve,
It is; but only within that medium. The 'colour' per definition doesn't change.
Hubert ('Lupus') remarks, https://www.mikroskopie-forum.de/index.php?topic=40111.msg297027#msg297027 (https://www.mikroskopie-forum.de/index.php?topic=40111.msg297027#msg297027)
Zitat
Der Vorteil einer kürzeren Beleuchtungswellenlänge (z.B. LED mit 365 nm) ist ein doppelter: Einerseits nimmt die beugungstheoretische Auflösung des Objektives im Wellenlängenverhältnis zu. Andererseits vergrößert sich auch die Phasenverschiebung durch die optische Weglänge innerhalb der Objektstruktur relativ zur jetzt kürzeren Wellenlänge, und damit verstärkt sich im gleichen Maß der Bildkontrast. Das gilt speziell auch für die normale Hellfeldbeleuchtung.
The reverse is then also true: if the contrast increases by higher RI, then also the wavelength is reduced. Within that medium. Chicken and egg tale.
As for increased resolution (lateral), I don't find any conclusive evidence for that. The images from
Amphipleura pellucida in that thread are all very well, but possibly only witout the exception of the images from Peter H with UV illumination, I doubt any of them would fulfill the Rayleigh criterium for proper separation of the pores (including mine).
Looking forward to discussion!
Best wishes, René
Zitat von: Beatsy in Februar 26, 2023, 23:12:27 NACHMITTAGS
Last week, while reading an old thread, I saw a comment stating that high RI mountants reduce illumination wavelength
Hallo Steve,
Das tut es, aber nur innerhalb dieses Mediums. Per Definition ändert sich die "Farbe" nicht.
Hubert ('Lupus') merkt an, https://www.mikroskopie-forum.de/index.php?topic=40111.msg297027#msg297027 (https://www.mikroskopie-forum.de/index.php?topic=40111.msg297027#msg297027)
Zitat
Der Vorteil einer kürzeren Beleuchtungswellenlänge (z.B. LED bei 365 nm) ist ein zweifacher: Zum einen steigt die beugungstheoretische Auflösung des Objektivs mit dem Wellenlängenverhältnis. Zum anderen nimmt die Phasenverschiebung aufgrund der optischen Weglänge innerhalb der Objektstruktur in Bezug auf die nun kürzere Wellenlänge zu, was zu einer entsprechenden Erhöhung des Bildkontrasts führt. Dies gilt insbesondere für die normale Hellfeldbeleuchtung.
Umgekehrt gilt also auch: Wenn der Kontrast durch einen höheren RI-Wert zunimmt, dann nimmt auch die Wellenlänge ab. Innerhalb dieses Mediums. Die Geschichte vom Huhn und Ei.
Was die erhöhte (laterale) Auflösung angeht, so finde ich keine schlüssigen Beweise dafür. Die Bilder von
Amphipleura pellucida in diesem Thread sind alle sehr gut, aber möglicherweise nur mit Ausnahme der Bilder von Peter H mit UV-Beleuchtung, ich bezweifle, dass irgendeines von ihnen das Rayleigh-Kriterium für eine korrekte Trennung der Poren erfüllen würde.
Ich freue mich auf die Diskussion!
Beste Grüße, René
Übersetzt mit www.DeepL.com/Translator (kostenlose Version)
Hallo zusammen,
vielleicht hilft dieser Artikel aus dem Mikrokosmos von 2009 etwas.
http://www.mikroskopie-ph.de/Mikrokosmos,%20Band%2098%282%29,%20S.%2098%20%282009%29%20Brechwertmessungen%20an%20Diatomeen.pdf
Die Messungen waren relativ aufwändig, aber haben sich sicher gelohnt.
Viele Grüße
Peter
Hello Steve,
the problem is rather complex, one could write a book about it.
ZitatI always thought high RI mountants do not improve resolution but only increase contrast.
Resolution and contrast cannot be separated, the Rayleigh criterion is itself based on an arbitrary definition in which contrast is important for the separation of two objects. You also have to take into account the observation system, e.g. the specific properties of the eye or the camera sensor. For example, pixel noise affects the contrast and thus also the achievable resolution.
The biggest problem, however, is that we are dealing here with phase objects, where the usual mathematical formulas for resolution - whether according to Rayleigh or Abbe - apply only to a very limited extent. The formulas assume idealized amplitude objects in the form of points or lines. In the case of more complex phase objects, the illumination method has a very strong influence on the detectability of the details, but also produces different artifacts. As an example of this kind of problem, I have mounted Michael's comparison with different illumination methods
https://www.mikroskopie-forum.de/index.php?topic=40111.msg296776;topicseen#msg296776 (https://www.mikroskopie-forum.de/index.php?topic=40111.msg296776;topicseen#msg296776)
directly beside each other including a SEM image taken by Wilfried https://www.mikroskopie-forum.de/index.php?topic=28656.0 (https://www.mikroskopie-forum.de/index.php?topic=28656.0), which shows the real structure.
If you want to make a competition who is photographing the smallest structure with the highest contrast that's fine. But a serious documentation of the real structure of diatoms cannot be achieved by exhausting all possible contrast enhancement methods, because you can only interpret these pictures if you know before how they should look like.
If the object is embedded in an environment with a stronger refractive index difference, then a phase contrast method does not have to act as strongly to produce sufficient intensity changes. Then even normal brightfield can be sufficient for good contrast and higher resolution. And the pixel noise is less disturbing because usually a lower contrast enhancement is performed during image processing. This is an indirect effect that also affects the resolution. Contrary, if the refractive index difference is too large, a phase contrast method can even be disadvantageous, because the phase shift of the structure to be resolved is too strong and is no longer converted into intensity according to the object.
ZitatBut the formulae for calculating axial resolution (in z) do include RI as a term and axial resolution is increased by higher refractive index mountants
I don't think that the theoretically changed axial resolution is relevant for the structural resolution of diatoms. At the resolution limit of optics, only abrupt phase changes can be made visible, like openings in the diatom shell. I agree with Rene's opinion that the images in the linked thread do not demonstrate clear increases in resolution, the imaging techniques are too different for that.
Hubert
Thanks everyone for your responses and the interesting links.
I should have qualified my comment about the images I linked to "proving" more resolution at higher RI. It is correct that they are nearly all taken in different situations and can prove nothing conclusively. But combined with my own observations, I still think there might be something there. I am likely mistaken, but need to prove it for myself...
I understand that contrast techniques compromise resolution too. I use various contrast-enhancing methods when observing diatoms by eye - darkfield, DIC, oblique and (rarely) phase contrast - but when I photograph diatoms, I use bright field almost exclusively now. I rely on zrax (RI 1.72) for better contrast and the high dynamic range of modern camera sensors to stretch that contrast further in post-processing.
I am waiting for my camera to be returned after conversion for UV imaging and will try relevant comparisons under more controlled conditions when I get it back. I will, of course, share my results here.
Thanks again for all the thoughtful responses.
Cheers
Beats
Hello Steve
just one more addition, a medium with a higher refractive index simply put leads to two changes: the wavelength is shortened because the speed of light is slower in the medium. at the same time, however, the na gets worse (in the medium) and compensates exactly for the shorter wavelength. the resolution does not change! the contrast, however, improves due to the increase in the difference between the object and the medium. the mtf improves, which is +- the same as an increase in resolution. the effect of the medium with a higher refractive index is obvious, i have a preparation with amphipleura pellucida in yellow medium (2.3) the details can be seen visually without any problems in white light.
Martin
Thanks Martin - useful information. I'd still like to discover if axial resolution (which does improve at higher RI) has any other visible effect. It probably won't - but I need to see that for myself or it will bug me :)
I am intrigued by your "yellow medium". RI 2.3 is practically off the scale for a mountant! Is it realgar (also known as 'ruby of arsenic')?
..yes it is realgar, arsenic in an ancient preparation, in the blue the medium is opaque and therefore not usable. let us know if you have any new findings on axial resolution.
martin
Hallo,
vor Jahren hat mir Klaus Kemp ein Sonderpräparat angefertigt. Heute sicher nicht möglich und Klaus lebt leider auch nichtmehr.
Gruß
Peter
Lieber Peter,
ein Immersionsmedium mit n=1,9??? Hast Du irgendeine Information über deren Natur? Ich halte das für höchst unwahrscheinlich. Die einzigen fluiden Medien in diesem Bereich die ich kenne, sind Lösungen von Arsentribromid in Dijodmethan und die sind eklig, unbeständig und sehr stark gefärbt. Schmelzen von Selen und Schwefel gehen von diesem Bereich bis weit über n=2, aber die kristallisieren schnell aus.
Herzliche Grüße, Olaf
Lieber Olaf,
im Jahr 2008 hatte ich mehrfach lockeren Kontakt mit K.D. Kemp.
Er hatte mir geschrieben, dass er durch einen Nachlass eine kleine Menge Realgar, " gelbes Medium" bekommen hat. Abgeben wollte er auf keinen Fall, ( kann ich verstehen ) aber ich konnte ihn dazu bewegen mir ein Amphipleura p. Präparat mit diesem Gift anzufertigen.
Mehr kann ich dazu nicht berichten. Ob Anne mehr dazu berichten kann?
Schöne Grüße
Peter
Lieber Peter,
leider nein, die ganzen Chemikalien von Klaus Kemp hat der Quekett Club übernommen.
lg
anne
Das mit dem Realgar hat mich jetzt interessiert. Anscheinend wird das im Vakuum aufgedampft:
https://onlinelibrary.wiley.com/doi/abs/10.1111/j.1365-2818.1940.tb00867.x
Viele Grüsse
Florian
Hallo Florian,
in MIKROKOSMOS 1985 S. 55 - 59 ist von A. Meller eine ganze Menge beschrieben.
Die wichtigen Punkte sind von Seite 58 - 59. Ich würde nicht empfehlen die Chemikalien am Küchentisch zu rühren. Zudem sind reichlich Nachteile verbunden.
Gruß
Peter
Zitat von: peter-h in März 02, 2023, 16:49:25 NACHMITTAGS
Hallo Florian,
in MIKROKOSMOS 1985 S. 55 - 59 ist von A. Meller eine ganze Menge beschrieben.
Die wichtigen Punkte sind von Seite 58 - 59. Ich würde nicht empfehlen die Chemikalien am Küchentisch zu rühren. Zudem sind reichlich Nachteile verbunden.
Gruß
Peter
Hallo Peter,
vielen Dank, den Artikel hatte ich schon mal gesehen, wirklich übles Zeug. Demgegenüber erscheint die Vakuumbedampfung mit Realgar fast harmlos. Realgar ist ja auf jeder Mineralienbörse praktisch in Reinform verfügbar. Wäre ein heroisches Bastelprojekt.
Viele Grüsse
Florian
Lieber Florian,
was soll denn die Bedampfung mit Realgar (AsS) überhaupt bringen? Damit erhöht man doch nur scheinbar den Brechungsindex der Diatomee. Eigentlich ist man ja aber auf der Suche nach einem Immersionsmittel mit hohem Brechungsindex.
Herzliche Grüße,
Olaf
Lieber Olaf,
wie ich das verstanden habe, wird das Realgar aufsublimiert, aber in genügend dicker Schicht um als Immersionsmedium zu fungieren.
Viele Grüsse
Florian
... glaub' ich nicht, man müßte dann ja eine planparallele Schicht erzeugen, damit das ganze funktioniert. Ich laß mich aber belehren! (und bitte der Realgar - definitiv nicht genderfähig) 8).
Herzliche Grüße,
Olaf
Hallo Florian,
Du hast doch sicher die Anleitung gelesen. ;)
Gruß
Peter
Zitat von: olaf.med in März 03, 2023, 10:25:21 VORMITTAG
... glaub' ich nicht, man müßte dann ja eine planparallele Schicht erzeugen, damit das ganze funktioniert. Ich laß mich aber belehren! (und bitte der Realgar - definitiv nicht genderfähig) 8).
Herzliche Grüße,
Olaf
Lieber Olaf,
hier ist beschrieben, wie man den Realgar aufsublimiert. Die Diatomeen befinden sich dabei wohlgemerkt auf dem Deckglas, siehe hier, oder unten:
https://www.forgottenbooks.com/en/download/TheAmericanMonthlyMicroscopicalJournal1885_10568386.pdf
Auf den Artikel mit der Vakuumsublimationsanlage habe ich leider keinen Vollzugriff.
Jedenfalls muss man nicht vorher eine Lösung in Arsentribromid herstellen.
Viele Grüsse
Florian
Exc el l e nt ,
and ev e n b ette r , m o unts , as to per
man enc e , may be made by using
re algar only by subl imation . A bit
of the realgar is put on a plate of
mica abo ut one inch square , and
thick as a p e nny . This is m elted
by strong h eat of a spirit
- lamp .
On this mica plate is place d an
oth e r , w ith a ho l e 5 of an inch
in di a m e te r , and abo ve this a thin
glass plate with a h ol e sl ightly l e ss
than the glass co ve r on w hich the
diatoms are
m o unt e d .
In F ig . 2 5,
a and 6 are
the two li1 i
ca plates , 6 the glass plate , and d the
cov e r , with the diato msfacing the real
gar . The w hol e is now supp o rted on a
me tal ring . A stro ng h eat wi ll volati
l iz e the re algar w ith out change , and
ofte n a cl ear deposit is mad e all ove r
the diatoms and unde rsid e of the
cove r , and the latte r can n ow be
mounted in balsam ; but ,
if bubble s
a re fo rm ed in the op e rati o n , as prob
ably w i l l be the case ,
the h eat must
be co ntinu ed ti ll th e se disapp ear ,
and , as the deposit w i ll now be
thick e st at the c e ntre just ove r the
re algar , the mount may be fi nished
by putting the cov e r , re algar side
down , o n a cl ean sl id e , and 0 11 top
of it , to prev e nt bre aking, a pi e ce
o f thick glass , and th e n , gi asping
tightly with forc eps to gi ve pressure ,
h eating strongly ove r a spirit lamp .
The re algar w i ll softe n (it must not
be m elted , e lse bubbl es will form
which cann ot be re m oved)
and
spre ad out . mo re or l e ss , be twee n
the c ov e r and sl id e , making a nic e
cl ear m ount . The co lo r of the
h eated realgar is v e ry much deep e r
than wh e n cold .
Lieber Florian,
das ist tatsächlich ein Einbetten in eine Schmelze und nicht das was ich unter Bedampfen verstehe. Klingt abenteuerlich - vielleicht versuchst Du Dich einmal daran.
Herzliche Grüße, Olaf