Radiation detector types/physics

[artis]

While watching the MIT online opencourse videos about ionizing radiation physics some questions came to mind.
Let me first clarify and reinforce some basics by asking them to you. I understand this might be long, please forgive me. I will appreciate your time.

1) A GM tube consists of a chamber of low pressure gas and electrodes kept at some potential (can be changed) an incoming charged particle or photon above the threshold energy can cause a local ionization event that then gets to an avalanche cascade with the help of the E field potential across the gas gap between the electrodes?2)Can a GM detector work in both pulse and current mode? Say for low CPM activity it can count individual ionization events as the resulting current pulse from each individual ionization but as the activity increases and the events become closer spaced there develops a steady current in the tube? Can this limit were the pulse mode goes into steady current mode be controlled by changing the Electric potential between the electrodes , so say lowering the potential would allow the pulse mode to be extended to a higher activity?
Can the potential between the electrodes in a GM tube be considered a type of gain like that in an amplifier, where by increasing the potential results in a more sensitive detector in case for individual low near threshold photons/charged particles?
3) Would it be fair to say that a semicondutor detector is somewhat similar to a GM tube whereby it also has two electrodes with a potential applied across them but the material in which the detection process takes place is different? Where in a GM tube you would have low pressure gas which can be ionized by an incoming particle , while in a semiconductor detector like the Germanium ones there is a semiconductor between the electrodes and an incoming photon for example interacts with an electron by scattering it and the electron then with multiple other electrons excited by the primary electron get deflected towards the anode due to the E field potential across the semiconductor?
Here is the part that bugs me. With respect to spectroscopy.
I can understand why a GM tube can't give spectroscopy type results because it can only interpret radiation as electrical current and I suspect that one can get the same resulting current from a lower energy but higher intensity radiation as well as from a higher energy lower intensity radiation, but isn't this the same problem in the semiconductor or scintillator type detectors?
Because as far as I know irrespective of the detector type all eventually convert the radiation to electric current, so why would the current produced by a say 1.4MeV gamma interacting by scattering or photo electron emission within a semiconductor detector be rather precisely measurable but not within a GM tube if the end result from that gamma interaction is just a pulse of current?

Or is it that in a semiconductor detector due to materials used (low work function) and cooling to minimize background/electrical noise one can simply discern much lower radiation energies than one could in a GM tube if used in a pulse mode with the goal of determining the incoming particle energy?

 

[dRic2]

I'm not an expert, I only studied in a course a year ago. Anyway.

1) yes
2) mhm I don't think they are used in current mode because they have a pretty high dead time, but I don't know.
3) yes

In GM, since you have a huge electric field, the avalanched produced is not proportional anymore to the energy of the incoming radiation, but any particle will produce the same signal (regardless of its energy). It's like you reach a sort of saturation. You can work with smaller electric field and preserve the proportionality, in this case you are dealing with proportional counters or ion chambers. For those detectors (like for semiconductors) the number of charge carriers created is still proportional to the energy of the incoming radiation so the pulse that you get is indeed correlated to the energy and you can do spectroscopy.

 

[mfb]

Can the potential between the electrodes in a GM tube be considered a type of gain like that in an amplifier, where by increasing the potential results in a more sensitive detector in case for individual low near threshold photons/charged particles?

Yes. GM operation has a "gain" so high that your HV goes down, stopping the discharge. Very easy to read out but it comes with deadtime. Lower the voltage and you can get into the proportional region.
Not sure if there is an application where you want to transition between these regions. They won't have the same optimal geometry, gas mixture and so on.

3) Would it be fair to say that a semicondutor detector is somewhat similar to a GM tube whereby it also has two electrodes with a potential applied across them but the material in which the detection process takes place is different?

The analogy isn't very good. Normally you don't have the same amplification process. Avalanche semiconductor detectors are a thing but that's not your typical semiconductor detector where you only collect the produced charges.

With respect to spectroscopy.

That's a completely different topic.
GM gives you the same signal independent of the incoming radiation - anything that starts the chain reaction leads to the same signal. This has nothing to do with the intensity.
Semiconductor detectors give you a signal that's proportional to the deposited energy. If the particle is stopped in the detector that's the whole energy, otherwise it depends on the velocity (and type) of the particle. Proportional gas detectors are sensitive to the ionization from the particle, so they measure the velocity as well. All these measure individual particles. One at a time.

 

[gleem]

Regarding gas filled chamber type detectors of which the GM is one. If you decrease the voltage across the collecting volume sufficiently ( 300 V) you can create a constant current proportional to the rate of ionization occurring in the chamber. This is good for higher intensity radiation sources of the order of 10⁷ Bq or higher. The current and thus the calibration will depend on the specific ionization of the radiation, the volume exposed, and the pressure of the gas within. Of course, there is no energy information. Such a detector is usually used to measure exposure rates from an x-ray/gamma-ray source.

 

[artis]

1) @mfb @gleem I think I understand why GM detector cannot measure particle energy, because the moment radiation causes ionization in the gas inside the tube there starts a current flow between the electrodes so what the detector is actually "reading" is it's own power supply being shorted out with a variable resistance (the gas breakdown inside the tube). So there is no/can't be no information gathered about the original incoming particle energy because that is only the "kickstarter" of the current but the current is a result of the tube's power supply.

So the GM tube can work in a pulse mode if it measures low activity source?
Is there some sort of current limiting when the Gm tube measures high activity as dose? Because high activity would cause a constant (maxed out? ) current through the tube? This question assume the tube works in constant current mode (if there is such in typical GM detectors? )

Could this be the reason why some of the less capable dose measuring meters at Chernobyl maxed out, because the radiation was so intense that there was a constant short circuit within the tube?
2) With respect to semiconductor detectors and say for example a 1.460 MeV gamma ray from K40,
So say the incoming gamma ray produced a pair within the detector material (electron positron) this pair then quickly annihilates within the detector producing two gammas of energy the same as of the particles, so two 511KeV gammas, right?
If both of these 511KeV gammas manage to escape the detector (double escape) then the detector eventually doesn't capture any of this interaction that just took place within it?
Is it because even though a pair is made from electron positron but they manage to destroy themselves before they can interfere with the electrodes or the doped semiconductor material? 3) What if just one gamma escapes and the other gamma manages to get Compton scattered or photoelectric effect? Does then the liberated electron show up at the electrodes and get's counted and it's energy that gets shown on the display is the energy of the gamma that created it (511KeV) minus the energy required to "push" it towards the electrode?
I assume the energy within semiconductor detector to move an electron into the conduction band is very low (couple of eV?) ?

Finally , can a semiconductor detector actually count a single particle interaction if what I described here above say would be the only particle that hit the detector in a considerable time? Or is there a lower limit to the minimum amount of particles that have to be at certain energy to then be registered as a noticeable peak over the background noise?

thanks.

 

[artis]

Oh by the way , when I was talking about the incoming 1460KeV gamma , it produces a pair but the energy of that pair is 511KeV x2 so 1.022MeV, where does the rest leftover energy between the 1.022MeV and the 1.460MeV gamma go?

 

[gleem]

Could this be the reason why some of the less capable dose measuring meters at Chernobyl maxed out, because the radiation was so intense that there was a constant short circuit within the tube?

Yes. This is one hazard of using a GM detector (ratemeter) in a high intensity radiation field. The dead time becomes so large that it begins to miss counts and the rate that is registered decreases even though the radiation level increases. Eventually, if the radiation becomes too high and the dead time makes the detector insensitive to the radiation.

In referring to a GM counter you are talking about a gas filled chamber detector that is running at a certain voltage with a special gas. Such a detector has no current mode. However, if you replace this gas mixture with air and reduce the voltage to about 300V this configuration is called an ionization chamber (detector). The current produced depends on the volume and pressure of the gas. The current is small of the order of nanoamps per R/sec for about a one cc chamber at atmospheric pressure. So you can measure quite high intensity radiation field.

Oh by the way , when I was talking about the incoming 1460KeV gamma , it produces a pair but the energy of that pair is 511KeV x2 so 1.022MeV, where does the rest leftover energy between the 1.022MeV and the 1.460MeV gamma go?

Think about it. The gamma is converted to masses that are, what? (moving)

BTW when you use a GM counter to estimate dose rates you need to know the calibration factor for that energy gamma or mixture of gammas since the sensitivity significantly depends on the energy.

 

[mfb]

If the photon produces an electron/positron pair in your detector then you get electron/hole pairs from these two particles moving through your detector material. That's what semiconductor detectors measure, after all. If they don't slow down and annihilate directly they don't produce 511 keV photons - they'll have a higher energy. That process is too rare to be relevant, especially as the same process can happen outside the sensitive material as well with much fewer constraints.

So the GM tube can work in a pulse mode if it measures low activity source?
Is there some sort of current limiting when the Gm tube measures high activity as dose?

Yes.

Finally , can a semiconductor detector actually count a single particle interaction if what I described here above say would be the only particle that hit the detector in a considerable time? Or is there a lower limit to the minimum amount of particles that have to be at certain energy to then be registered as a noticeable peak over the background noise?

This is a question a 30 second google search will answer definitively. Or just think about what a tracking detector does.

3) What if just one gamma escapes and the other gamma manages to get Compton scattered or photoelectric effect? Does then the liberated electron show up at the electrodes and get's counted and it's energy that gets shown on the display is the energy of the gamma that created it (511KeV) minus the energy required to "push" it towards the electrode?

If you want to measure the total energy of a particle then of course you need to make sure all its energy is deposited in your active material. Semiconductor detectors are the wrong material to measure all the energy of high energy photons.

 

[artis]

@gleem OK right so the excess photon mass in a produced pair simply goes to the pair's kinetic energy , is this true for any photon energy, say a photon with 50 times more energy than the 1.022MeV needed to produce a pair will still produce a single pair and the rs of it's "leftover" energy will be added to the pair's KE?

@mfb so your saying that the produced pair from say a 1.460MeV gamma can instead of annihilation right there "on the spot" just drift some distance? But isn't that distance like very small given the pair consists of charged particles that have mass ?
So the higher the gamma energy the larger the possibility and further the distance a created pair will drift within a material where it was produced?
When speaking about holes, it's just an electron that left the place and then another electron eventually goes to fill it , but the hole itself doesn't go anywhere right? So eventually it's electrons moving in one direction (like in a diode which I read the semiconductor detector is) where one got liberated and struck the anode and another one filled it's place releasing a photon while doing so?
Not being lazy or anything but I did much more than a 30 second google search, maybe the wrong keywords but I couldn't find the answer.
My own guess would be that a semiconductor detector if well isolated could count a single interaction within it's medium , is that true?
as for your last reply, I think I know why , because a semiconductor detector is a "thin" device so even though it's SNR is good and it's sensitive is high , most high energy gammas fly through without interaction and of those that do many interact just partly like the single escape case,
this is also the reason why CERN in it's detectors use layers upon layers of absorber material mixed with scintillators etc to capture the particles while measuring their energy decrease at the same time, would this be a correct reasoning ?

 

[mfb]

@gleem OK right so the excess photon mass in a produced pair simply goes to the pair's kinetic energy , is this true for any photon energy, say a photon with 50 times more energy than the 1.022MeV needed to produce a pair will still produce a single pair and the rs of it's "leftover" energy will be added to the pair's KE?

That's the most common result.

@mfb so your saying that the produced pair from say a 1.460MeV gamma can instead of annihilation right there "on the spot" just drift some distance? But isn't that distance like very small given the pair consists of charged particles that have mass ?

You produce two charged particles at high speed flying in a somewhat different direction. What exactly do you expect them to do, if not flying through the material?
Higher energy particles tend to have a larger range, yes.

Holes are an extremely useful concept in semiconductors. Don't try to avoid it.

My own guess would be that a semiconductor detector if well isolated could count a single interaction within it's medium , is that true?

A charged particle doesn't do a single interaction, it does many, typically hundreds of them that lead to electron/hole pairs. They are easy to detect with a high efficiency.

CERN detectors don't even try to find objects in the low MeV range. They look for GeV particles, where photons and electrons produce electromagnetic showers in the calorimeters.

 

[artis]

Ok guys , I'm back with a question.
I read these following links and more
https://www.nrc.gov/docs/ML1122/ML11229A683.pdf
https://www.psi.ch/sites/default/files/import/lmu/DevPsdEN/semicond_detect_review.pdf

So semiconductor detectors come in various forms, the P-N junction diode types, surface barrier ones, PIP , where in all of them some form of P or N type substrate is employed with additional dopants and electrode metals.
Then there are the shall I say more exotic SiLi and GeLi of which the latter is superseded by the HPGE due to it's better performance I think.But here is something I don't understand. In the diode type detectors for example, whether they are used as is or reverse biased, say I want analyze the spectrum of alphas/betas, (from what I read the thin diode type detectors are mostly used for aphas/betas and low energy gammas)
So the diode detector gives off current due to the radiation interaction within it. This current is then sent to a preamplifier. But if I have multiple alphas with different energies hitting the diode, is it not the case that it will "detect" only the particle with the highest energy?
I ask this because if multiple particles with different energies are hitting the diode/detector simultaneously and from what I know multiple parallel currents add together then isn't it the case that the output will show just the largest current corresponding with the highest incoming energy particle/s ?
Like in this reverse biased photodiode counter , please see the link, page 3.
https://inis.iaea.org/collection/NCLCollectionStore/_Public/49/018/49018135.pdfThe same question I would ask about the HPGE detector for gamma spectroscopy.
Again say we have a source that emits multiple gamma energies , but radiation is emitted constantly so I suppose the different gammas are emitted simultaneously and they hit the detector and overlap , so how can the detector Ge crystal and it's connected electronics discriminate/differentiate between the multiple energies if for a high intensity source there is a constant current running through the detector the magnitude of which is directly proportional to the incoming gammas with the highest energy?Are there some "tricks" used as for example sampling where the detector output is taken in discrete steps and by such one can catch the moment of both high energy gammas (larger current) and lower energy ones (lower current), or is the radiation itself discrete and the time delay between each gamma big enough for the detector current to be able to change in time so that a current proportional to each incoming gamma can be recorded? thanks.

 

[gleem]

If two particles in fact arrive at the same time or during the "live time" of the detector their energies will be summed. To avoid this the intensity of the source must be decreased to make such a situation unlikely.

 

[artis]

@gleem So it would be fair to say that a semiconductor detector of any sort is incapable of spectroscopy if the activity is so high that on average multiple particles hit the detector at the same time ?

A quick google search seems to suggest that for a Ge detector the CPS limit is about 10k , so that would indicate a detector recovery time of about 100 μs after each particle interaction?Is it true that scintillation detectors on average count faster so can analyze a hotter sample but have a lower resolution than semiconductor types due to the higher energy needed for detectable interaction within the detector?

 

[mfb]

Semiconductor detectors used for tracking are faster than tens of nanoseconds. Otherwise ATLAS and CMS would struggle keeping the different events apart (25 ns separation). LHCb has that issue with its outer tracker where they need to consider multiple events as hits might take a while to be detected - but that's a gas detector, not a semiconductor design.

@gleem So it would be fair to say that a semiconductor detector of any sort is incapable of spectroscopy if the activity is so high that on average multiple particles hit the detector at the same time ?

That's generally true everywhere. Reduce the activity of the source, increase the distance, make the detector more granular, use a faster detector, ...

If you want specific advice about specific detectors, you could at least tell us which radiation and energy range you are interested in. You asked about 1.4 MeV gammas before but you wouldn't use semiconductors for that.

 

[artis]

@mfb LHCb outer detector is gas, so is it the straw tube type one? Where the thin long tubes are filled with gas and have a central electrode made of a fine thin wire?

I wasn't after specific advice since I'm overlooking multiple detector systems as per the thread title.
Ok I see , that is what I thought , all the detectors work in the dose mode if activity is so high that events overlap.
So basically for a given detector a hotter sample just needs to be kept further away for spectroscopy and problem solved, at least for gammas , not sure about alphas where increasing the distance might lose some of the lower energy ones alltogether?Why you say I wouldn't use semi detector for a 1.4MeV gammas? I recently watched a MIT opencourse video where they did the classical experiment of frying up a bunch of bananas and producing banana ash, which they then squeezed into a plastic bag and put inside a cooled and isolated chamber that had a HPGE detector barrel at the bottom of it.
PS. while we are at it, I just thought about this. Various crystals and organic substances produce scintillation light due to interaction with high energy particles. If I were to put a fluorescent tube inside or close to a high activity source would it light up? In theory it should as the electron energy levels shouldn't care about whether they are excited by electric current or random charged particles or photons.?

 

[gleem]

So basically for a given detector a hotter sample just needs to be kept further away for spectroscopy and problem solved, at least for gammas , not sure about alphas where increasing the distance might lose some of the lower energy ones alltogether?

Place the alpha source and detector in a vacuum chamber.

 

[mfb]

Things being too active for an energy measurement is very rare.

Why you say I wouldn't use semi detector for a 1.4MeV gammas?

How do you make sure to get the full energy deposited in your detector reasonably often?

If I were to put a fluorescent tube inside or close to a high activity source would it light up?

If the source is strong enough (and it's not alphas), it should produce some light.

 

[artis]

How do you make sure to get the full energy deposited in your detector reasonably often?

Well I'm not sure what you mean by "reasonably often" , I guess one would need to leave the detector "running" for some hours of time in order to get a full picture of what is in the banana ash , well we know what is in there but to make it fully visible in the spectrum ?
can you please elaborate on your objection as it is not entirely clear to me , especially given I've seen they do this in other places?

Now one argument could be that a smaller semi detector is very thin from a gamma perspective so most of them fly right through , although I see those detectors within a chamber that has special shielding where the gammas might also interact from within with the shielding producing pairs where the photons from the annihilated pair then hit the detector either one of them or both and then produce an interaction so this "reflect" radiation might help in terms of measurement time?

 

[artis]

Oh and while we are at it , one specific question with regards to the Super Kamiokande detector.
Now normal scintillator PMT's work by having the PAMT tube and at the end there is a photocathode which has a low work function and so after that is a scintillation crystal or material and when incoming radiation interacts within the material it produces photons in the visible spectrum that then hit the photocathode and eject electrons that then get accelerated/multiplied in the dynodes of the PMT to produce a current pulse. Now in the Super Kamiokande , if what I understand is correct then the PMT tubes are themselves without a scintillation material attached and are essentially transparent with no photocathode at the end ?
The scintillation material is itself the column of purified water and the PMT's just surround that water tank wall serving as "pixels" so if an incident neutrino passes through producing light within the water then that light would directly pass onto the closest PMT'S first and furthest later and produce a corresponding current and then by judging the current pulses and creation times they would analyze and come up with a neutrino trajectory in the water etc ?

 

[mfb]

I guess one would need to leave the detector "running" for some hours of time in order to get a full picture of what is in the banana ash

You'll never find everything.

especially given I've seen they do this in other places?

Do what? Measure that there is radiation? Measure some very rough energy estimate for some of its radiation? Do a precision measurement of K-40 decay energies? These are completely different tasks.

"Reasonably often" is not just a matter of statistics. It's also a matter of having a peak that's clearly distinct from the remaining events where you measure a lower energy (or background from elsewhere).

although I see those detectors within a chamber that has special shielding where the gammas might also interact from within with the shielding producing pairs where the photons from the annihilated pair then hit the detector either one of them or both and then produce an interaction so this "reflect" radiation might help in terms of measurement time?

That doesn't give you the right energy. It measures the presence of radiation, sure, but see above, different goals.

Now in the Super Kamiokande , if what I understand is correct then the PMT tubes are themselves without a scintillation material attached and are essentially transparent with no photocathode at the end ?
The scintillation material is itself the column of purified water and the PMT's just surround that water tank wall serving as "pixels" so if an incident neutrino passes through producing light within the water then that light would directly pass onto the closest PMT'S first and furthest later and produce a corresponding current and then by judging the current pulses and creation times they would analyze and come up with a neutrino trajectory in the water etc ?

The neutrino doesn't produce light, but secondary particles do if the neutrino interacts. By measuring the arrival position and time it's possible to reconstruct their tracks.

 

[artis]

@mfb Ok I agree about the bananas or the K40 in them, measuring like they did in the MIT video would give you a rough estimate of what element or material you are measuring by looking at the peaks given. So say I would then separate the K40 chemically (like Cody did in a youtube video for example) , what kind of detector and setup would then be best to use to get the very precise decay energies considering in most cases it decays by an electron and the rest by the already mentioned 1.460MeV gamma?As for the kamiokande experiment. Well ok I think I now understand how it works. Can you please tell me what information would it give us about neutrinos if we saw their presence by a secondary means (their interactions with water atoms inside the purified water tank) ? Would we try to estimate their mass based on the energy given to secondary particles as well as the original neutrino trakjectory vs the secondary particle ones?
I presume we are after neutrino mass which seems to eluded us so far as it seems to be tiny based on the fact that they normally escape detectors without depositing any energy in them, this must also be the reason why the detector is built underground (to shield against other forms of radiation) and has a huge tank of water increasing the chance of interaction?

Wiki says the flux of solar neutrinos is 65 billion per square cm per second which sounds huge ,

 

[bobob]

@mfb Ok I agree about the bananas or the K40 in them, measuring like they did in the MIT video would give you a rough estimate of what element or material you are measuring by looking at the peaks given. So say I would then separate the K40 chemically (like Cody did in a youtube video for example) , what kind of detector and setup would then be best to use to get the very precise decay energies considering in most cases it decays by an electron and the rest by the already mentioned 1.460MeV gamma?

A Compton supressed HPGe detector would work just fine for a $\gamma$ in that energy range. HPGe detectors can be very thick - as much as 10 cm thick. (The geometry is co-axial, not planar.) You surround the HPGe detector with, for example, scintillators to detect a $\gamma$ that was Compton scattered out of the HPGe and you only accept events where you get a signal in the HPGe and NO signal from the scintillator. You can get energy resolution of a couple of keV for a 1MeV $\gamma$.

As for the kamiokande experiment. Well ok I think I now understand how it works. Can you please tell me what information would it give us about neutrinos if we saw their presence by a secondary means (their interactions with water atoms inside the purified water tank) ? Would we try to estimate their mass based on the energy given to secondary particles as well as the original neutrino trakjectory vs the secondary particle ones?

The wikipedia article explains this in some detail. The short version is that you get ring of cerenkov radiation projected on the phototubes. The opening angle of the cerenkov radiation is inversely proportional to the velocity of the particle that produced it -- in this case, an electron, a muon or a tau. They infer whether it is an electron or muon based on the sharpness of the cerenkov ring. The electron is light and multiple scatters more than a muon so the ring of light from the muon has sharper edges.

I presume we are after neutrino mass which seems to eluded us so far as it seems to be tiny based on the fact that they normally escape detectors without depositing any energy in them, this must also be the reason why the detector is built underground (to shield against other forms of radiation) and has a huge tank of water increasing the chance of interaction?

Well, the electron neutrino has to be very light or you would notice it in the endpoint energy from $\beta$ decay spectra. The $\beta$ endpoint is the point where the electron carries off the maximum energy available to it. The energy of the decay is shared between the electron and neutrino. See, for example:

https://arxiv.org/pdf/0909.2104

 

[artis]

thanks so far @mfb (although I suppose you kind of got tired of my questions) and @bobob
One more thing, since neutrinos interact very rarely , is it true that it is especially hard to detect slow neutrinos the ones with low KE, because in most experiments their presence is known from the secondary phenomena they produce during their interaction so in a photomultiplier tube experiment like the Kamiokande the neutrinos have to produce Cherenkov light in order to be detected at all, but in order to produce the cherenkov they have to be with a high KE so one could say there is a high cutoff limit for their minimum detectable energy?

 

[bobob]

thanks so far @mfb (although I suppose you kind of got tired of my questions) and @bobob
One more thing, since neutrinos interact very rarely , is it true that it is especially hard to detect slow neutrinos the ones with low KE, because in most experiments their presence is known from the secondary phenomena they produce during their interaction so in a photomultiplier tube experiment like the Kamiokande the neutrinos have to produce Cherenkov light in order to be detected at all, but in order to produce the cherenkov they have to be with a high KE so one could say there is a high cutoff limit for their minimum detectable energy?

The neutrinos don't produce Cerenkov radiation. Cerenkov radiation is produced by charged particles. The Cerenkov light then comes from the interaction of the neutrino with the electrons in the water and the light produced by the electron or the light produced by secondary reactions from the electron that was produced. The cross section is proportional to the energy, so the higher the neutrino energy, the greater the chance of an interaction with an electron. Low energy neutrinos are harder to detect because the cross section for the interaction is smaller. Also, the neutrinos need to have enough energy to transfer to the electron such that you get Cerenkov radiation from the electron (or outgoing muon or tau).

 

[artis]

@bobob well thanks for the clarification although I knew that they themselves don't produce cherenkov radiation that is why I said "secondary phenomena".
When I asked about the energy threshold is because the electron KE/speed has to be higher than the speed of light in water for it to shine so a neutrino that say excites an electron below that energy won't be detected in a detector like the Kamiokande. So I assume this must be the lower energy detection limit for neutrinos in such a detector.