Safe viewing of laser and UV light through optical scope

[LaserGuy]

I am using an optical microscope for the assembly of some small parts, and that assembly involves using a red laser (class 3R) and a UV lamp. Both light sources will be pointed perpendicular to, or away from, the optical path of the microscope, but obviously there's room for error which could be bad if some of the harmful rays enter the microscope and are magnified into my eyes. So I plan to mitigate that using filters for the appropriate wavelengths. I've found filters which would fit into the eye pieces of the scope and attenuate the light to 3% of its original intensity.

The problem is I can't seem to find any information online about how to assess the risk of using lasers/UV with microscopes so that I can determine if 97% attenuation is enough. There are plenty of statements saying that "They can pose severe eye hazards when viewed through optical instruments (e.g., microscopes, binoculars, or other collecting optics)", but how to properly assess and mitigate the hazard? Presumably the intensity at the eye is a function of the incident/scattered light into the microscope (wavelength, intensity, beam profile, angle of incidence), the microscope optics, and the eye itself. But with the lack of information I'm seeing online it seems that my options are a) get a PhD in optics and do that analysis from scratch, which could go very badly if I make incorrect assumptions and end up wrong by orders of magnitude, or b) assume that the risk is too high and can't be mitigated with filters and instead look into building some kind of shutter interlock to completely block the light path whenever the hazardous source is activated.

If anyone has expertise in this area and could point me in the right direction, even just key words to search, I would greatly appreciate it!

[Sure Squintsalot]

I have a similar question: how to safely view episcopically illuminated subjects with (relatively high intensity) UV light?

I suppose viewing the LCD display while live-viewing would work with either UV or laser, but that seems awfully impractical. Then there's using laser protecting glasses, but that too looks impractical.

What do the laser confocal people do?

[PeteM]

The laser confocal people image directly to camera. This is an option for the original question as well - doing the viewing on a monitor.

Most UV microscopes have highly effective UV blocking filters - and also shields around the stage to handle stray reflections. Plain (and cheap) clear polycarbonate shields as used for welding helmets are said to block near 99% of harmful UV (link below). So, that material can be a cheap source of added assurance.

It does bear more investigation - the effect of magnification, allowances for a bit of puffery - and outright fraud in some cases. I recall buying UV filters off eBay from an overseas supplier. When tested with a spectrometer, they appeared to be pretty much clear glass.

https://www.mcrsafety.com/~/media/mcrsa ... letter.pdf

[Sure Squintsalot]

I figured the fluoresence set-ups have prisms that do triple duty in reflecting UV down to the sample, passing specific wavelegths up, and blocking all UV to the eyepieces. Absent a purpose-built fluorescence set-up, I'd have to find where to put the UV filter before it gets to the eyepiece.

Then there's the issues of objectives. None of this works without the right objectives.

[PeteM]

Typical fluorescence scope of the sort we hobbyists might use has a "cube" that passes the excitation frequency down to the sample and a blocking filter that passes the emission frequency but blocks UV - pretty much as you've said. These cubes have a half mirror (reflects light down, passes light back up) and positions for up to three filters.

In addition, there's often a permanently installed UV blocking filter located above the cube area that provides assurance in the event a BF or DF cube is moved into position, a filter degrades or cracks, etc. I'd think a DIY setup should surely have something at least equivalent to a polycarbonate UV filter in that position and better yet a purpose-built barrier filter of optical quality.

Scopes will also have a solid shutter of some sort for the UV source.

[Sure Squintsalot]

In addition, there's often a permanently installed UV blocking filter located above the cube area that provides assurance in the event a BF or DF cube is moved into position, a filter degrades or cracks, etc. I'd think a DIY setup should surely have something at least equivalent to a polycarbonate UV filter in that position and better yet a purpose-built barrier filter of optical quality.

Scopes will also have a solid shutter of some sort for the UV source.

OK. THIS is good to know.

[Alexander]

A properly set up fluorescence scope will never let any UV into the eye-pieces. your application does not sound like being in the fluorescence business, more like reflected light microscopy. In this case a filter blocking UV in the light path is what you need. There are filters blocking everything below 400 nm and above 700 nm letting pass only the visible light. They are found in astronomy shops. Another option is a simple pol-filter. They block UV. To attenuate all light is not the best of all ideas. You may get a very dark and weak picture.

I would not scare myself about the stray light. Sun-bathing on a California beach exposes you to amounts of UV that are at least 100 times more intensive. The danger comes from the eye-pieces. If there is UV ....

[GerryR]

I know from a little research in using UV for sterilization purposes, that the spectrum is basically divided into 3 parts, UVA, UVB, and UVC, with UVC being the most harmful of the three and best for sterilization. You might learn more about UV if you research Germicidal Sterilization equipment as they provide a list of precautions, and how to protect yourself, skin and eyes, when using it. Just a thought.

I found this "white paper" with some info:

https://www.christiedigital.com/globala ... paper1.pdf

[Alexander]

UVC would be blocked by the glass in Objective and eye-piece. UVC stray isn't a huge problem either, because it is absorbed by air even on short distances and has very limited capabilities to penetrate skin.

[LaserGuy]

I should have stated in my original post that this is for reflective microscopy. So there will potentially be light reflected off of the work area and, in worst case scenario of losing control of the optical fibre (635 nm laser) or light wand (UV lamp), light may accidentally enter directly into the scope at full intensity. We're talking 9+ W/cm^2 UVA, 1.5 W/cm^2 UVB, and a 2.5 mW red laser.

I appreciate all the input and I understand that using filters is probably safe enough. But in a setting where I'm responsible for the safety of others I'm hoping I can definitively prove, either analytically or experimentally, that the transmitted light is not dangerous for the operator.

Some reading tells me that the worst case scenario for the operator is a perfectly collimated beam entering the eye, and of course the smaller the spot size the higher the intensity. So for an analytical answer, I imagine I would need to construct a scenario where the beam is as collimated as reasonably possible, and the spot size at the output of the microscope is as small as reasonably possible in this setup. If I were well versed in lasers and microscope optics then maybe I would be able to create a simply model for this worst case scenario, but I'm not familiar enough to be able to make the assumptions for that model.

For the experimental proof, I was thinking that maybe I could direct each light source into the objective and then, with and without the filters, measure the power and spot size at the eyepiece to determine the intensity [mW/cm^2]. But that doesn't account for additional focusing by the user's eye, which would be worst case if the beam was perfectly collimated so I guess I would need to measure for divergence/convergence as well.

[JMC]

I'll come back to this thread with a bit more info when I have more time (currently laid up with Covid and spending 20 hours a day sleeping).

I built a uv transmission microscope and eye safety was a key concern. I use filters below the eyepieves which block >99.99% of the light below 400nm. 97% doesn't sound good enough to me. I also measured the irradiance spectra at the eyepieces - that is the only way to be certain, and with eyes and uv, I'd want to be certain.

The red laser adds another complexity. Perhaps something like a 10nm or 40nm bandpass filter centered around 500nm would be useful. That would block uv and red light. See if Thorlabs has anything suitable. I'd be looking at OD4 blocking (<0.01% transmission of unwanted wavelengths).

I do not use the eyepieces when imaging with uv (focus check is done with live view), and I always wear uv safety glasses when the light source is on. Safety is layered on my setup. Only take something going wrong once to produce major eye damage.

[Scarodactyl]

If I were you I would not DIY this.

[Alexander]

If I were you I would not DIY this.

Given that

But in a setting where I'm responsible for the safety of others I'm hoping I can definitively prove, either analytically or experimentally, that the transmitted light is not dangerous for the operator.

I fully support your statement not to do it DIY.

[LaserGuy]

I don't intend for this to be "DIY" quality, or to have any doubt. Like I said, I will only go ahead if I can achieve a strong safety justification. Based on feedback on here, and the reading I've been doing, I don't think I will get there so will most likely go for a viewing screen or an interlocked shutter.

[crb5]

The hazards of uv light and visible laser light involve different physical principles.
Low power laser light is potentially dangerous because the near parallel beam can be brought to focus at the back of the eye and the local high intensity can cause damage to the retina. Lasers in 3R class (visible wavelengths and 1 to 5 mW power) are deemed safe for use as laser pointers because the eye can respond by blinking and turning away, such that any blindness is temporary. The caveat about looking through optical instruments is often stated, but with little or no explanation. As I understand it, the danger comes from expanding the beam diameter from 1 mm to around 7 mm (the maximum diameter of the pupil). The intensity of the beam per unit area is reduced but the eye lens could focus the 7 mm beam to a smaller spot and hence a higher intensity at the retina (equivalent to high NA lenses giving better resolution). In theory, the diameter of the focused spot would be reduced about 7 fold (and hence intensity could be 50 fold higher).

https://www.holoor.co.il/optical-calcul ... spot-size/

Thus, it would not be a good idea to look directly at a laser pointer through 7x binoculars. The danger comes from beam expanding optics. Passing a visible laser beam through a microscope objective lens, as in a confocal microscope, causes the beam to focus on the sample a few millimeters away, but then the beam diverges and poses no danger to the operator positioned a few feet away. In the typical epi illumination used in microscopy, the laser beam is reflected towards the sample and would only pose a danger if the operator chose to view and focus on a plane mirror which would reflect the light back through the optical path. If the filter cube contained a blocking filter with an OD > 2 (blocking 99%) even this scenario should be rendered safe with a <5mW laser source.

uv light poses a different problem because the photons are high energy regardless of whether the beam is focused or not. Such light does not penetrate the eye well, so damage mainly occurs at the cornea. The really dangerous wavelengths (<320 nm) are blocked by glass (hence the need to use silica or quartz for instruments operating at these wavelengths). Furthermore, many of the specialist glasses used in microscope objectives block in the 320-400 nm range (and hence the need for specialist fluar objectives for wavelengths around 350 nm). The biggest danger from uv light comes when the source is operated detached from the microscope, as may be required when initially centering and collimating an arc lamp in its housing. Here the beam is shone on a wall a few meters away. The operator should wear uv blocking glasses for this process. Otherwise, as pointed out earlier, once assembled onto the microscope with a uv blocking filter in the epi-cube, a uv light source is less dangerous than a day at the beach or on the snow slopes without sunglasses.

Just a few caveats

Some laser pointers are labelled <5mW but exceed this value by several fold. Be suspicious of a bright 405 nm pointers as the eye is less sensitive to this wavelength so it should not appear as bright as a 450 nm pointer of the same power rating.

Some visible lasers (not HeNe) generate light by doubling infrared frequencies. A filter is used to block the IR light, but it may be removed by rogue operators to get higher intensities.

When using lasers with a microscope, first hold a piece of white card at an angle over the eyepiece to see if there is a bright beam coming through before attempting to look directly by eye. A card is a very useful accessory when dealing with low to medium power lasers for checking the beam path, since once you know where the beam is going, the remaining space is safe.

I do not have a PhD in physics but have used medium power lasers (5 to 500 mW) in the lab for 30 years. I would be happy to receive comments/corrections as it is never too late to learn.

[Scarodactyl]

I don't intend for this to be "DIY" quality, or to have any doubt.

I think it could probably be done perfectly safely, but in a setting like this it might be better to have the protection of a vetted system so blame can't fall on you in that way if someone manages to misuse the system in such a way as to get hurt