Dabbling with fluorescence on a budget

[Charles Krebs]

Off and on I have thought of jumping into fluorescence microscopy. I've gathered together the bits and pieces needed to set up one of my Olympus biological BHS stands, but never got enough ambition to make up the light sources needed. (Kudos to Pau for his wonderful solution: http://www.photomacrography.net/forum/v ... hp?t=32736 ) I often thought about about Nightsea's elegantly simple system and how something similar would be a natural fit for my MM-11 hybrid scope.

https://www.nightsea.com/products/stere ... e-adapter/
https://www.youtube.com/watch?v=_oGXbXY9MU0

When I came across some relatively inexpensive flashlights with Nichia 365nm UV LEDs I decided to give it a try. These are the flashlights:



They were purchased from this site:
http://www.gearbest.com/led-flashlights/pp_277704.html

I liked the fact that these used rechargeable 18650 Lithium ion batteries, as I have some of these and they can power a light like this for a good long photo session.

The other piece I needed was a longpass or "barrier filter". This was purchased from Thorlabs here:
https://www.thorlabs.com/newgrouppage9. ... oup_id=999
I purchased both a FGL420 and a FGL435. So far I have only used the FGL420.

The filter was placed in the filter drawer above the objectives as indicated by the blue arrow below.



The flashlight is held by a small articulating arm. (I use mini Arca type clamps on all my articulated arms). The long thin shape of the light allows it to be positioned close to the subject.



I have a regular white light directed on the subject from the other side. I set up the shot using the white light. I also set up the automated stack start and end points with the white light. When I am ready to go I shut off the white light and turn on the UV. This is all new territory for me but the first couple of efforts seems to show that this can be a very viable approach. (And my cost for the light and filter was less than $60).
First shots (and some additional discussion) here: http://www.photomacrography.net/forum/v ... hp?t=33112

My wife tells me I worry about too many things ... but I strongly suggest you take appropriate precautions when working with UV light. I use black foamcore panels to prevent any overspill off the microscope stage. I also use a readily placed and removed piece it as a "shield" between my eyes and the microscope stage. I also have UV protective "wrap-around" safety glasses (Uvex Genesis XC model) purchased here (pretty inexpensive):
http://www.safetyglassesusa.com/
Maybe all this is overdoing the concern. I don't know, but I really don't care. It is not much effort to protect your eyesight.

Pau had asked which Nichia LED this is. I can't be certain but I believe is is the "NVSU233A U365" seen on this page:
http://www.nichia.co.jp/en/product/uvled.html

You can see a violet color from the LED. I think this is supposed to be the spectral output:



I am using the LED flashlight without an excitation filter. Based on the graph above and the detailed specifications of the FGL420 it looks like I might get away without using one. If I determine that I am cutting it too close I'll try either the FGL435 longpass "barrier" or use a filter on the light (from Ebay seller bjomejag) that sharply blocks anything longer than 390nm.

[Pau]

Charles, thanks for posting, your setup is closer to Jacek's, that is also working very nicely.
http://www.photomacrography.net/forum/v ... 870#201870

I'll need to try it in the future, now I'm convinced that it's better for low magnification than the traditional EPI Fl I'm using
It seems that you don't need excitation filters, In case of doubt you can try Schott UG1 or UG5, or an interference one
I have three UV excitation filters, all three work well although not equally well:
- Schott UG1, similar to UG5, form an old Zeiss PM: it works well although it eats some excitation light. I allows to pass some red light but this is no problem with UV LEDs
- Chroma AT350/50x: it is clearly the best one
- Omega 330WB70 from bjomejag site, it eats more excitation light than the Chroma, maybe even more than the Schott, likely because it is centered at shorter wavelength

[johan]

Another filter that works very well as a UV-pass filter is the Baader Venus filter. My lord, it's eye-wateringly pricey, though.

I'm hoping to try a slightly different approach, which is to cut up a prototype torch that was made for me and make a thing out of the cut up remainder connected to a lab power supply that I can interchange on the epi component in place of the regular light housing. Would anyone have a contact in the UK, south of London, that could do such cutting work for me?

[ChrisR]

A ZWB2 ( or other ZWB) "black" filter for the torch/flashlight would be good. Unfortunately all the listings I've looked at so far have them as no longer in production. They were only a couple of $.

[zzffnn]

Thank you for your fine efforts, Charles. Those are very helpful to me.

I have seen your great fluorescence images, shot using that UV flashlight rig. Are you getting enough chlorophyll fluorescence for video recording, using the same rig? Thanks again.

[banania]

Thanks Charles for posting your very interesting low budget solution. The Nichia you are using seems very usable and practical and cheap. I just ordered one, maybe I should have ordered two. Are you using Nichias at both sides or just one side when shooting.

Your Nichia is only 3W but powerful as it is LED and can be held very close to the subject and the spectral output looks also very nice and narrow and there seem to be very little visible light involved. I gather that exposure times and ISO can be kept nicely short/low even with one lamp?

I am afraid that my expensive Raytech UV lamp just went obsolete, at least the LW part. I have been using it for macros up to 2X and short stacks also. It has both LW and SW tubes and filters , 6W each, but it is very big and cumbersome and requires pushing up both ISO and exposure times, like ISO 800 and 30s, meaning very long stacking times and a lot of noise.


Henri

[Charles Krebs]

harisA posted a nice look (and some measurements) of a couple of UV LEDs. Worth a look:
http://www.photomacrography.net/forum/v ... hp?t=33134

And remember... you can't see 365nm UV light (or anthing shorter than about 390nm). The violet color you see with the Nichia LED is likely just that small trailing off seen in the graph between 390nm and 410nm. You can't see the vast majority of the output.
PROTECT YOUR EYES!

banania:
<<. Are you using Nichias at both sides or just one side when shooting. >>
So far just the one. But it does appear that there will be advantages with two with some 3-dimensional subjects and at certain magnifications.

zzffrnn:
<<Are you getting enough chlorophyll fluorescence for video recording, using the same rig? >>

I'm really new to this! It obviously varies greatly with the subject and the brightness of the objective (see: http://www.olympusmicro.com/primer/phot ... rrors.html )

Also from the little I know 365nm may not be the best excitation wavelength for what you want to see... perhaps something longer like ~450nm might prove better. It is actually a pretty complex subject (just Google something like: "chlorophyll auto fluorescence, best excitation wavelength" and you will see what I mean.) Hopefully those with more experience than I have will add some information.

What I can do is provide the exposure information used for the few shots I've posted. (right now I don't have access to the images,... I will later tonight. Check back late tonight or tomorrow and I'll add it to this post.

_________________________________________

OK Fan, referencing pictures in post http://www.photomacrography.net/forum/v ... hp?t=33112

1) ISO 200 1/2 sec, 10/0.28 objective
2) ISO 200 2 sec, 10/0.28 objective
3) ISO 200 1.3 sec, 20/0.40 objective
4) ISO 400 1/8 sec, 10/0.28

The shot posted here: http://www.photomacrography.net/forum/v ... hp?t=33133

ISO 400 2 sec, 50/0.50 objective

All were done with a single light.

[jcb]

For what it's worthI have been doing close-ups of chlorophylle autoluminescence in algae (green, red and brown) using :

• an electronic flash ;
  2 layers of blue acetate used in theater lighting (Lee 721 : peak transmission near 430 nm) attached with tape on the flash ;
  1 red filter on the camera lens (Heliopan R25 : no transmission below 570nm)

Superimposing the blue and red filter blocks all light.
That was the cheapest I could find that worked. Of course you have to work in the dark.

[zzffnn]

Thank you very much, Charles and jcb.

Charles,
It sounds like my video recording will not work, based on those exposure parameters.

jcb,
Lee #363 is cleaner than #721, but it is also less efficient.

[johan]

JCB, would you have a few images to point me towards using that filter combination?

[ChrisR]

By the way - something others may be able to comment on:
one wouldn't want to use something excitable like white paper as a UV reflector. I have read that the dull side of Al foil works well for "fill" light.

[zzffnn]

I bought this 420nm filter from eBay just now, for $16 shipped:
http://www.ebay.com/itm/262634372466?_t ... EBIDX%3AIT

It appears to be highly similar to, if not the same as, the Thorlabs FGL420 filter that Charles uses. I don't know if Thorlabs charges shipping. So buying from eBay will save $9 + Thorlabs shipping fees.

[ChrisR]

Poking around the internet, there are pieces of work using 405nm lasers as excitation for Chlorophyll A and B.
These (and LEDs at the same nm), are much easier to obtain than, and of course not as hazardous as true UV sources. Nice narrow spectrum but narrow beam too - how well can one diffuse a laser?

[banania]

I have tried to photograph fluorescence with 405nm laser pointers (not real lasers!) and I don't think they are suitable for photography purposes.

I tried a 100mw 405nm pointer, which is actually really dangerous. The light bounces from reflective surfaces and you would not want the beam to hit your eye!. Actually 100mw laser pens are forbidden (consumer use) here in Finland, but some chinese vendors sell 100mw pointers with a 5mw sticker so that it would not be caught at the customs. I bought mine from Russia where there are no rules for these. I then tried also a 5mw 405nm pointer but even that weaker pointer created very strong pointlike illumination so that the image was otherwise dark but had a much overexposed area where the beam hit. As it is only 5mw diffusing the light suitably would produce a very faint 405nm illumination. The pointer is however quite useful in checking possible fluorescence very fast and it is widely used for example to find out if crystals have been repaired etc...

There are also lots of led flashlights available with about 400nm peak light. These are often sold as scorpion flashes. I bought a couple of those also. One was with 50 leds but for fluorescence purposes they are not very good as the spectral peak is fairly wide.

The LED Charles suggested is the first cheap and viable alternative to dedicated UV lamps, like Raytech models, to use in photography (that I know of). I searhced for these a couple of years ago, but didn't find anything usable, so bought a Raytech.

I did some UV photography yesterday and was going through the options for a background that would yield just a tiny amount of visible light through fluorescence. I was thinking about tin foil and even cut a piece to try it, but I was únsure about its reflective properties and decided against it. As the reflected light is invisible there is not any easy way to figure out how much of it is reflected by the foil. If it has been tried and found OK for the purpose it would be nice here more about it. Maybe it is OK for 365nm UV photography when adequately protected with glasses etc,, but it would be nice to read something about it as I shoot mostly with 254nm light.


Henri

[enricosavazzi]

I don't know if it has already been posted, but if you need an excitation filter to remove the VIS blue and violet at the tail of an UV LED emission spectrum, there is no need to use one of the modern (and expensive) filters for UV imaging on the UV torch. It is enough to use one of the much cheaper legacy filters (U-340 or equivalent, 2 mm thick, is best for a 365 nm LED).

Modern filters for UV imaging on digital cameras are expensive because they must block NIR with an OD of at least 3 or 4, in addition to VIS. Legacy filters like U-340, U-360, UG11 etc. all transmit way too much red and/or NIR for using on a "full spectrum" digital camera. There is no need for this when the purpose is to remove the blue/violet "leak" of a UV LED, because the latter emits no NIR.