Diatom imaged with 313nm light

[jmc]

Over the last couple of years I've been building a UV transmission microscope to help with my research (some of my earlier post are here - viewtopic.php?f=8&t=41772). I recently had a diatom slide made with a few different species on it. The slide and coverslip are quartz so don't block the UV light.

One of the benefits of shorter wavelengths is increased resolution for a given NA. This is a bright field image taken at 313nm and shows the central portion of one of the diatoms on the slide. I've reduced the resolution for sharing on here as the original is 7360x4912.

Going in close to a cropped part of the original image gave the following.

The 'round' features are about 500nm across and the gaps between them about 200nm.

The objective used was a Leitz 100x NA 1.20 UV glycerine immersion one (see below), and the condenser was an antique Zeiss quartz one.

Here is a lower magnification image taken at 313nm with a Zeiss 32x Ultrafluar objective, showing the whole diatom.

[Macro_Cosmos]

Lovely results! Where did you purchase the quartz coverslips from? I have a source that offers sapphire coverslips and at nearly $60 a pop, it is just too expensive.

Here is mine with blue light.

https://live.staticflickr.com/65535/519 ... 4671_o.jpg

[jmc]

Thanks Macro_Cosmos. I got them from UQG Optics in the UK, and they were custom made. I think they were around 20GBP each. I built the microscope for my work, so things like quartz slides and coverslips I can buy as work related items. They'd be too expensive if it wasn't work related.

Lovely image of yours. I'm certainly no expert at imaging diatoms, but enjoy taking the photos. They are such beautiful structures and something I'll do more with in the future.

[rjlittlefield]

This is lovely, and working very well!

I like to check the numbers periodically, to reinforce my understanding.

With illumination at 313 nm (in air), and NA 1.2, the spatial frequency diffraction cutoff (MTF=0) should be at lambda/(2*NA) = 130 nm per cycle. Assuming perfect optics, MTF=39% will happen at twice that scale, around 260 nm per cycle.

In your image, I measure good contrast with feature spacings of 250-300 nm per cycle, so the image looks exactly as the math says it should. Nice!

--Rik

[jmc]

Rik, thanks, I tend to use Abbe's equation as you have (lambda/2NA) so yes, I was expecting the resolution you stated. However my objective and condenser weren't the same NA and I have seen other approaches which take that into account, so I take these numbers as guidance and tend to rely on what I see in the images.

I'd like to get hold of a calibrated test chart slide at some point. I have found some uv fused silica ones which have lines down to about 100nm but I am currently making my mind up whether or not to invest in one as they are very expensive.

[Vincent Zhang-1968]

It's great, thanks for sharing!

[jmc]

It's great, thanks for sharing!

No problem, you're welcome.

[Lou Jost]

Nice results! If anyone is interested in other objectives for this application, Nikon makes a "UV-F" series of glycerine-immersion objectives in 10x, 20x, 40x, and 100x, the latter two with NA of 1.30. All have exceptionally high NA for their magnifications; the 10x has an NA of 0.50. They are for 160mm tubes. I have some of these but have not yet tried them.

There is a complete set sitting on eBay for quite a while:
https://www.ebay.com/itm/303281617325?h ... SwH5tdeQpS

The 100x comes up often, and is inexpensive. The 40x comes up less often.

I think these might work with UV epi lighting without need for exotic materials or special condensers. The cover slip itself does not block much UV, if I am remembering your experimental results correctly. The light would have to cross the slip twice, but maybe even that would not be too bad, at least for those of us who can't find quartz condensers.

Edit: I've just found a source that measured UV transmission of these and found that they transmit only above 325-330nm.

[ModelZ]

Good work, thanks for posting!
Yours is interestingly just a hair below UVA. Never done that, so can't comment how abruptly things change. But e.g. for narrow band 365 nm excitation (above 340 or so) you often do not need any special optics. For example Reichert (Plan or not) Achros work fine there and all fluorites should work above 320. They do not produce noticeable autofluorescence. Zeiss Neofluars work, too (but at least the Z. Apos I've tried don't). Same with R. Neo DF condensor and older Leitz cardioids, they all work pretty efficiently in that UVA window.
-Karl

[Lou Jost]

I just found this interesting article suggesting that diatom frustules interact differently with UV light versus visible light. This might make UV light more effective for photography than we might expect by the wavelength-resolution advantage?
https://www.nature.com/articles/s41598-018-21810-2

[jmc]

I just found this interesting article suggesting that diatom frustules interact differently with UV light versus visible light. This might make UV light more effective for photography than we might expect by the wavelength-resolution advantage?
https://www.nature.com/articles/s41598-018-21810-2

Oooh, interesting, thank you for sharing Lou.

[Beatsy]

I just found this interesting article suggesting that diatom frustules interact differently with UV light versus visible light. This might make UV light more effective for photography than we might expect by the wavelength-resolution advantage?
https://www.nature.com/articles/s41598-018-21810-2

Interesting indeed! This quote from Darwin (in Origin of Species) has always stuck in my mind...

"Few objects are more beautiful than the minute siliceous cases of the diatomaceae: were these created that they might be examined and admired under the higher powers of the microscope? The beauty in this latter case, and in many others, is apparently wholly due to symmetry of growth..."

I've never fathomed an answer to that, but this paper might go some way to explaining it. Thanks for the link.

And nice images Jonathan! I'm getting a UV itch 'cos of you

[jmc]

I just found this interesting article suggesting that diatom frustules interact differently with UV light versus visible light. This might make UV light more effective for photography than we might expect by the wavelength-resolution advantage?
https://www.nature.com/articles/s41598-018-21810-2

Interesting indeed! This quote from Darwin (in Origin of Species) has always stuck in my mind...

"Few objects are more beautiful than the minute siliceous cases of the diatomaceae: were these created that they might be examined and admired under the higher powers of the microscope? The beauty in this latter case, and in many others, is apparently wholly due to symmetry of growth..."

I've never fathomed an answer to that, but this paper might go some way to explaining it. Thanks for the link.

And nice images Jonathan! I'm getting a UV itch 'cos of you

Cheers Beatsy. After years of skin and sunscreen research, when someone says 'I'm getting a UV itch', it's normally a bad thing, but I get what you mean
I'd love the go to shorter wavelengths and see what they look like, but that will have to wait for now - 313nm is as low as I can realistically go with my current equipment.

[René]

I'm not at all convinced by this report, typical tunnelvision of a couple of material scientists. But indeed Lou, instead of using a high refractive mountant to create contrast (which doesn't work in UV in any case), you can use water and still be able to generate (absorbance) contrast. As usual for most subjects, and certainly within the Abbe diffraction limits.

Thanks Jonathan, for showing this structure of Pleurosigma. Glad to see Roderich Zeiss saw the right thing some hundred+ years ago ;-) https://www.alamy.com/the-microscope-an ... d0%26pl%3d

Best wishes, René

[jmc]

I'm not at all convinced by this report, typical tunnelvision of a couple of material scientists. But indeed Lou, instead of using a high refractive mountant to create contrast (which doesn't work in UV in any case), you can use water and still be able to generate (absorbance) contrast. As usual for most subjects, and certainly within the Abbe diffraction limits.

Thanks Jonathan, for showing this structure of Pleurosigma. Glad to see Roderich Zeiss saw the right thing some hundred+ years ago ;-) https://www.alamy.com/the-microscope-an ... d0%26pl%3d

Best wishes, René

Thank you for sharing that - I shall find a copy of the book. I never cease to be amazed by the quality of the images that were being obtained over 100 years ago.