EPMA as a radiation hazard

[jon_wade]

Anyone had the joy of dealing with University safety offices regarding e-beam instruments as X-ray hazards?   Apparently, under UK 2017 legislation any e-beam equipment capable of >5kV accelerating voltages is a potential ionising radiation hazard, meaning all users need to be logged, tracked and deemed 'radiation workers'.  There is a bit of an exception for SEM's but, risk averse Safety Offices seem to not want to take the exception.  I'd be interested to hear your thoughts on this and if anyone else has experienced this (I have asked that all users use the departmental issue lead pants when entering the room...)

[Nicholas Ritchie]

At NIST, we have an ionizing radiation safety group that performs periodic inspections and certifies our instruments as safe.  The technician waves a GM over and around the instrument notes that there is no increase in count rate and issues a certificate which we hang on the instrument.  No additional requirements for logging and tracking users.  We do have radiation badges we wear for other purposes (When performing gamma spec for instance.)  The risk from 30 keV x-rays in a vacuum interlocked system is minimal even with aluminum plates although with one older instrument I was asked to add an additional layer of aluminum foil over a couple this aluminum plate ports.  I thought it a statistical fluke but decided it was better to just add a layer of aluminum foil than debate the point.  TEMs at 300 keV could produce X-rays that could escape under just the right conditions.  The best argument in your favor is the vacuum interlock.  The beam won't energize without vacuum.  When at vacuum, there must be sufficient mass-thickness contain 14 psi and this is more than enough to block a 30 keV X-ray.

[sem-geologist]

Getting into discussions if EPMA is a potential radiation hazard is like discussing if a glass of water is a potential drowning hazard. The later of course would be ridiculed just straight out by majority of population and thus we had not seen that happen, at least yet (although looking where our society is heading, I would not be surprised if in very close future we would be required to stick the labels by every sink and kitchen with "Drowning hazard!").

Vacuum interlock argument is kind of creative way to overcome idiocy, albeit it is like arguing that "glass of water is locked in a fridge - And so it is not possible to drown in it because of that".

But maybe we should still try to strive for more proper reasoned arguments? In that case we should focus on power. In case of XRF or XRD (which I agree has some potential air ionisation and thus has some hazard), X-ray tubes/lamps are used which use kilowatts of power to generate very strong electron stream which is only about 1% converted into X-rays. The presence of this huge power is obvious to anyone even without physics background -the water cooling of X-ray tubes requirement and very huge chillers to only cool the X-ray tube itself. None of EPMA's have water cooled E-guns (we have cooled coils, but that has nothing with radiation), because the total emitter power is normally not more 1W and often is much less - so only in this oversimplified glimpse the EPMA electron source is thousand times less poweful than in X-ray tubes. More over, only tiny fraction of generated electrons are accelerated to sample, so in the best case (i.e. Schottky emitters with enlarged tip surface) are capable to beam down no more than 1-5µA. That is million time less than what takes place in Xray lamps at most exaggerated situation. But in most of work we use tens of nanoAmperes - thus it is normally Billion times less than what Xray lamps beams down to the samples. And then the geometry: produced X-rays on the EPMA sample scatter all around (different to X-ray lamps where specific angle is designed to get most of X-rays be emitted mostly in one direction). So final value of X-ray intensity is diluted furthermore, so Even with WDS spectrometers Case off - the produced X-ray beam which hypothetically we could put our hand to is over few billion times less than what X-ray tube produce. And so considering air ionistaion in XRD machine and EPMA with WDS case off is Bilion time different values.

I think considering EPMA as radiation hazard illustrated with "glass of water as drowning hazard" is already very huge exaggeration by few orders of magnitude. More proper analogy for illustration would be "a water droplet as potential drowning hazard". If we consider EPMA as radiation hazard We should no more get near any sink as single water droplet could drown us...

[Probeman]

Getting into discussions if EPMA is a potential radiation hazard is like discussing if a glass of water is a potential drowning hazard. The later of course would be ridiculed just straight out by majority of population and thus we had not seen that happen, at least yet (although looking where our society is heading, I would not be surprised if in very close future we would be required to stick the labels by every sink and kitchen with "Drowning hazard!").

Exactly.

Just for fun I asked an AI about radiation from EPMA instruments operating at 30 keV:

Modern, properly shielded EPMA instruments do not emit high levels of radiation into the surrounding environment at 30 keV operation, and any leakage is typically at or near background levels. The instrument is designed to contain the X-rays generated during analysis within the lead-shielded column and chamber.
The radiation generated within the instrument includes:

    Characteristic X-rays: Produced when the high-energy electron beam (up to 30 keV) interacts with inner-shell electrons of the sample atoms, causing them to emit X-rays at energies characteristic of the elements present.
    Bremsstrahlung (continuum) X-rays: Generated as the incident electrons decelerate within the sample, producing a continuous spectrum of X-rays ranging in energy up to the full beam energy (30 keV).

These X-rays are used for the analysis itself and are detected by internal Wavelength Dispersive Spectrometers (WDS) and/or Energy Dispersive Spectrometers (EDS).

When asked what are typical doses measured within 1 meter from such instruments it replied:

The typical measured radiation within 1 meter of a properly shielded EPMA instrument in normal operation is negligible, often indistinguishable from the natural background radiation level. This level is generally less than 0.2 uSv/h (microsieverts per hour)

We can compare this value to the famous XKCD radiation chart seen here:



https://xkcd.com/radiation/

Which means that working within 1 meter of an EPMA instrument, your exposure is somewhere between eating a banana and living within 30 miles of a coal plant for one year. (hint most of your health concerns would be from mercury, arsenic and lead exposure!).

To provide more context, the radiation dose for an average individual is around 10 uSv per day (with a very large variance), while a single airplane flight from New York to Los Angeles, is around 40 uSv.

Maybe this chart should be printed out and posted in all microanalysis labs...?

[Nicholas Ritchie]

Of course you are both right about the lack of risk, but from a rad safety perspective it is the vacuum interlock that ensures that the beam isn't energized when the chamber is open.  This combined with the metal of the vacuum chamber is what ensures that you don't get a dose of 30 keV X-rays (which could in fact do some harm).

[Probeman]

Of course, though at a 30 keV electron beam one is not likely to generate many 30 keV characteristic or continuum x-rays (overvoltage and all that).

Interestingly, the Cameca instrument crystal alignment is performed on an operating instrument tuned to Fe Ka using column separation windows. You're not supposed to put your hand in the x-ray path when adjusting the crystal...   

Yes, the vaccuum interlock needs to be working, though if one did manage to open the airloock door at vaccuum you're going to have a significant implosion!  By the way this actually happened at UC Berkeley, due to some bad vacuum logic, just before I took over the EPMA lab on our old ARL SEMQ.  The mechanical and high voltage effects were impressive from what I heard and later saw.

I guess my point is just that a properly operating EPMA/SEM instrument produces about as much radiation as eating a banana, so if one is actually worried about the vacuum logic operating properly, then perhaps one should be testing those interlocks rather than performing a radiation measurement, which tells one very little on an properly operating instrument.

By the way, did they even turn on the high voltage when they made the radiation measurement?

[sem-geologist]

Interestingly, the Cameca instrument crystal alignment is performed on an operating instrument tuned to Fe Ka using column separation windows. You're not supposed to put your hand in the x-ray path when adjusting the crystal...   

It is irrelevant. The X-rays, few billion times weaker than from X-ray tubes (0.000015W vs 1000.000000W to 5000.000000W used in XRF or XRD), would practically be absorbed in its majority within surface of the lint-free glove (or rather double glove). Actually, I got an idea - I will do experiment by placing a single and double layer of nitrile glove in between XTAL and separation window, next time I will need to adjust the WDS and will share how much attenuation nitrile glove creates.

Vacuum interlock(s) on these machines are to protect the High Vacuum gauges, Ion pump(s), Diffusion pump or turbo pump, and of course electron emitter - their purpose is not to defend us from potential radiation hazard (which is not there, to begin with), but rather defend the machine from Us - defend from the user/operator and their mistakes, which can be catastrophic to the machine. Ironically, We posse more danger to machine than machine to us.

That of course does not change the fact that vacuum interlock argument still probably is one of most clever overuse and easiest to use wrong argument successfully defending against ignorant laws and enforcers. However, that is kind of playing their stupid games. You could also place a crystal ball in the room and tell the enforcers that it "sinks"/absorbs all the "bad" radiation from the room protecting the users, and ask to take G-M measurements to "prove" that it works. It is kind appalling that most important part – the "MICRO" – in electron MICROscopy and in electron MICROprobe is somehow ignored, which is crucial to grasp from just a glimpse (even with lack of any physics training) why they practically are powerless to pose any radiation hazard. Maybe due to overuse of "micro", "nano", "pico" unit prefixes as some catchy marketing blahberry in multitude of different kind of products had brainwashed masses so they don't see a difference between an ant and a whale anymore even if they are in plain sight?

[Probeman]

The X-rays, few billion times weaker than from X-ray tubes (0.000015W vs 1000.000000W to 5000.000000W used in XRF or XRD), would practically be absorbed in its majority within surface of the lint-free glove (or rather double glove).

I completely agree, but I thought it was funny that the safety people would probably freak out if they knew how we adjust our spectrometer crystals...

[JonF]

Anyone had the joy of dealing with University safety offices regarding e-beam instruments as X-ray hazards?  Apparently, under UK 2017 legislation any e-beam equipment capable of >5kV accelerating voltages is a potential ionising radiation hazard, meaning all users need to be logged, tracked and deemed 'radiation workers'.  There is a bit of an exception for SEM's but, risk averse Safety Offices seem to not want to take the exception.  I'd be interested to hear your thoughts on this and if anyone else has experienced this (I have asked that all users use the departmental issue lead pants when entering the room...)

The 2017 regulations that are being referred to are here: https://www.hse.gov.uk/pubns/books/l121.htm
It's worth giving them a read, as I find it helps to be on the same page as the H&S folk when dealing with any issues.

[Les Moore]

Reminds me of the safety concerns about methane in the exhaust P10 gas. My reply was that there was more coming from the operators than the machine :-)