How to measure the activity of a radioactive source

[dRic2]

TL;DR Summary

is measuring activity possible in current mode ?

My answer is "no".

Hi, I've been reading about radiation detectors (manly form Knoll's book), but there is something I don't understand. Radiation detectors are of very different nature, but they all share a common process to detect a type of radiation:
- I have a control volume
- The incoming radiation interacts with the material inside my control volume generating a "signal" (ionization or photon emission mainly)
- I collect whatever is generated from the interaction with an electric circuit
- The type of output current is then analyzed to get the information about the radiation

As I also understand some detector (like ion chambers) can work in current mode or in pulse mode (there is also a voltage mode, but I don't care about that right now).

The pulse mode is the most intuitive method: each quanta of radiation generates a signal. For example, if I have a $\alpha$ -emitter, each $\alpha$-particle generates a single "pulse" of current, so the activity of my source is simply obtained by counting the number of pulses during the measurement time. GM-counter or scintillators work this way, right ?

But there is a problem: measuring a single pulse of radiation takes time and while the detector is processing a single event it can not detect other events; so if the activity of my radioactive source is too high there will be too much event taking place inside the detector in a very short period of time. This will lead to a significant loss of events detected and thus a pulse mode cannot be applied if the source activity is too high. For this reason in such conditions it is preferred to operate the detector in current mode where the output consists of a time-averaged current. The simplest example I can think of are the ion chambers operated in current mode. My question is: if I work in current mode I can easily obtain the energy deposited in my detector (that is, I can easily calculate the dose), but how can I calculate the activity of the source ?

For example, consider again an $\alpha$-emitter, and suppose that, after the measurement, my ion chamber measures a value of current. From that value I can calculate the energy deposited by the $\alpha$-particles inside the ion-chamber by considering the energy required to generated a single ion-electron pair, but how can I tell whether that energy comes from a single high energetic $\alpha$-particle interacting several times or by a lot of less-energetic ones ? I think I can not.

 

[Paul Colby]

Why can't one relate total charge to count rate? Use a low level source to calibrate charge per pulse?

 

[dRic2]

calibrate charge per pulse

Yes, if you are working in pulse mode each pulse of charge is related to a single quanta. But in current mode you do not measure pulses: you only get a continuum flow of current. That flow is indeed proportional to the energy released in the detector but you have lost the information about the time of each pulse (ie, the single interaction).

Imagine this scenario:
you have an ion chamber and 3 alpha-particles interacts with the gas in a very short amount of time. If you work in current mode you don't measure 3 pulses, but a single longer signal. Now, that signal is obviously proportional to the energy deposited by the 3 particles, but you don't know that from the measurement. You only know that a certain amount of energy has been released. It could, for example, be due to a single alpha particle with higher energy interacting 3 times with the gas instead. You don't have enough information to discern between the two case because you lack the time-information about each singe interaction.

 

[Paul Colby]

you have an ion chamber and 3 alpha-particles interacts with the gas in a very short amount of time. If you work in current mode you don't measure 3 pulses, but a single longer signal. Now, that signal is obviously proportional to the energy deposited by the 3 particles, but you don't know that from the measurement. You only know that a certain amount of energy has been released. It could, for example, be due to a single alpha particle with higher energy interacting 3 times with the gas instead. You don't have enough information to discern between the two case because you lack the time-information about each singe interaction.

Okay, so you can still make statements about the energy per unit time deposited. If you require a detailed histogram of events binned by deposited energy you're out of luck. Depends on what you really require.

Now that said, one could still view the detector current as a signal with fluctuations in time. Perhaps one could glean information from its power spectrum? One would need the goals of the measurement and develop models of the current with time to make headway.

 

[dRic2]

One would need the goals of the measurement and develop models

Well, the goal is to measure activity, that is the event of decay per unit time. As I understand it is only possible if you register each single quanta of radiation. The information about energy might be useful to perform a spectroscopy (but you still need the information about the count rate) or if you are simply interested in the dose (that is the energy per unit mass deposited).

 

[Paul Colby]

but you still need the information about the count rate

What's wrong with a pile-up correction calculation?

 

[Paul Colby]

I still don't get your issue. If 3 alpha interact in a short period then one gets 3 times the charge in that period. What am I missing?

Okay, got it. If a weaker source of the same physics is measured, then the energy histogram is known. Is such a measurement possible?

 

[dRic2]

If a weaker source of the same physics is measured, then the energy histogram is known

I don't quite get what you mean by "weaker" and "same physics".

Anyway, If I understood correctly your method should work if I already know the type of source. For example, if I know that all the alpha particles carry the same energy and I know the value of that amount of energy, in principle I think you could relate the number of charged formed inside the detector with the number of particles interacting with it in that period of time.

Anyway if those particles have instead a very wide energy spectrum or if you simply don't know the kind of source you have (for example you have to characterize the sample) I think that it can not work.

 

[BvU]

Summary:: is measuring activity possible in current mode ?
My answer is "no".
I've been reading about radiation detectors (manly form Knoll's book)

You sure you are asking about activity ? That's a counting rate, not an energy.

If you have specific questions relating to a specific section in the book, it might be nice to indicate which sections you are refrring to.

 

[dRic2]

You sure you are asking about activity ? That's a counting rate, not an energy.

Exactly. That is the question. Since it's a counting rate I assume that it is not possible to measure it in current mode, but I'm not sure.

For example here are 4 pages taken from the book (I hope I can post them without problem)

and

As I understand the only way to measure activity in current mode is in the only particular case that the incoming particles are all known and at the same energy. But it is not explicitly stated in the book at any point, so I guess it's not the case or it is assumed obvious.

 

[BvU]

My hawkeye dsitinguished two blobs, one yellow -- before he talks about current mode

and one yellow-green -- after the section about current mode

I can't link these two to your

As I understand the only way to measure activity in current mode is in the only particular case that the incoming particles are all known and at the same energy. But it is not explicitly stated in the book at any point, so I guess it's not the case or it is assumed obvious.

but maybe I misunderstand something.

In your post #1 you mention ionization chambers as the subject of this thread, i.e. no gas multiplication mechanisms and such (as with geiger counters and proptubes) but operating in the ion chamber region.
So there is a pulse height distribution that smears the current vs counting rate relationship. That doesn't disqualify the instrument for use as activity meter, it just reduces the accuracy -- most of the time only a little, and for many purposes not problematic. You mention "incoming particles are all known and at the same energy" but even then the energy loss process is a statistical process -- not to mention geometrical considerations if it's not a pencil beam of particles.

Your scenario in post #3 is realistic in the sense that you should realize that a single particle has to ionize quite a lot of gas molecules for a signal to be detectable (a picoampere is $10^7$ ions/s) and the total number of ionisations for alpha particles that stop in the chamber is probably roughly proportional to the kinetic energy.
However, not many sources have such a spread in energy that the relationship current-activity is completely lost.

 

[dRic2]

Sorry if you had troubles reading my photos, if you want I can try to upload a better version

Maybe I'm poorly explaining my problem. Suppose you have an unknown source and you want to establish the activity of that source. Suppose you use an ionizing chamber in current mode. As the output of your measurement you get a current. How do you relate that current to the number of decays if you know nothing about the source ? The collected charge could have been originated by a few high energetic particle (thus low activity) or by lots of lower energetic particles (thus high activity). On the other end if you work in pulse mode you would have measured a number of pulses equal to the number of decay events, thus the calculation of the activity is easily carried out. This is kind of an extreme example, but suppose you have a known source that decays to a very short-life intermediate element which in turn decays emitting a different type of radiation (or the same one, but with a different energy). This is pretty common. Again, if you work in current mode how do you calibrate your ion-chamber to get the activity as an output ?

Current mode ion chambers seem to work only if I know "everything" about the source and if the energy-spread of the radiation is limited. In this way I can convert the output current accordingly to activity.

PS: Here we talked about ion chambers, but all of this apply to every detectors (like scintillators) with the exception of the GM counters. GM counters always work in pulse mode.

 

[Vanadium 50]

You seem to want to know many things from a single measurement - the decay rate, the total ionization, the type of radiation, the energy spectrum ("spectroscopy") etc. All with a single measurement. You aren't going to get this.

 

[anorlunda]

the activity of that source

Please define what you mean by that phrase.

 

[dRic2]

Please define what you mean by that phrase.

the number of decays per unit time

You seem to want to know many things from a single measurement - the decay rate, the total ionization, the type of radiation, the energy spectrum ("spectroscopy") etc. All with a single measurement. You aren't going to get this.

I only want to know the the number of decays per unit time, but with less information possible about the source. For example, I give you a sample of -let's say- $^{238}U$ and I'd like to know only the number of decays per unit time. My question is: do I operate the ion-chamber in pulse mode or in current mode ? Can I use both ? Why, why not ?
Now suppose I didn't tell you it was $^{238}U$ but somehow you know it is an alpha-emitter, do you operate the ion-chamber in current mode or in pulse mode ? Can you use both ? Why, why not ? (remember I only want to know the number of decays per unit time)

According to what I've said so far I could in principle use both methods for question 1 (although I'd prefer pulse mode) and I would be forced to work in pulse mode in the second scenario.

 

[Vanadium 50]

Do you actually have a device? The meter should be calibrated to counts per second, even if it's current mode.

It seems that you want counts per second, but don't want to use a dead-time correction (or more likely, a dead-time correction circuit). Further, you want it to give you an accurate count without knowing the kind of radiation you have.

That doesn't exist.

Extreme example why: tritium has a 6 KeV beta (average energy, up to 18 keV), which won't penetrate a standard survey meter tubes (we'll ignore specialty items for now). So you can look all day with a survey meter and see nothing.

A typical survey meter is insensitive to alphas. It is periodically calibrated to a known source. There is a check source on the device, so before using it, you check the meter gives the correct answer to a known source before using it.

 

[dRic2]

It seems that you want counts per second, but don't want to use a dead-time correction (or more likely, a dead-time correction circuit). Further, you want it to give you an accurate count without knowing the kind of radiation you have.

No, I am aware that you have to take into account the dead time, and the type of radiation is known, just the energy of the particles is unknown. A standard ion-chamber works perfectly fine for me: would you set it to current mode or to pulse mode to measure activity ? That's my question.

I remember once I was doing a Lab experience in university where we measured alpha particles (counts per second) coming from a sample of $^{238}U$ with an ion-chamber operating in pulse mode. I was wondering if I could have operated in current mode instead. Yeah, I should have started my post with this ahaha

 

[BvU]

Now he tells us ...

 

[dRic2]

Well, I was thinking in general terms, if I started with that example I thought I would get specific answers. The current mode/pulse mode regime is shared by most detectors, not only ion-chambers. The same question could also arise in a different experiment with a different detector. That is way I wanted to keep things as general as possible. Maybe there is a specific answer for that particular case...

Anyway I apologize for the inconvenient/misunderstanding

 

[BvU]

No need to apologize -- my sigh was meant ironically. We are in a learning environment where we can safely ask all kinds of questions (including ...) and give all kinds of answers (including ...). Gentle corrections aren't perceived as threatening.

I recall doing that Thoron lab too (in 1972 ) and I don't recall asking the kind of questions you brought up, so: kudos ! (All wasn't lost for me, fortunately: I did get a PhD in HEP eventually, perhaps because I never encountered ionization chambers ever since ).

Maybe someone in a teaching setting is better qualified to help you out, but 'general' answers must be considered for what they are.

 

[Vanadium 50]

I was wondering if I could have operated in current mode instead. Yeah, I should have started my post with this

It would have been good to let us know, yes.

So your question is essentially "I operated the detector in low rate mode. I was wondering if I could have operated it in high rate mode instead." (How "low-rate" and "high rate" are implemented is a detail) The answer, unfortunately, is "maybe". It depends on how high "low rate" is.

 

[dRic2]

@Vanadium 50 I'm not familiar with the low/high-rate terminology. Does it refer to the number of pulses ? Or to the number of decays ?

 

[Vanadium 50]

Pulse mode is intended for low rate. Current mode is intended for high rate.

 

[gleem]

The determination of the activity of a radioactive source is not easy and involves many steps particularly for an unknown source. Even with a known source and equipment that previously has been used with this source can be challenging depending on the sample.

For an unknown source which can be very challenging since you must first determine its half life, decay modes, and the decay modes of the daughters since all this radiation is produced at the same time. Any pulse can come from the parent or any of the daughters. Because of this, you need to select a detector, procedure, and assay configuration to unambiguously register only radiation from the parent decay or otherwise determining how to compensate for the daughter's decay products.

Typically ionization chambers are only used in current mode for relatively high radiation fluxes and for known sources. Ion chambers can be used to measure the activity based on the Specific Gamma Ray Dose Factor or Gamma Ray Constant. which is measured in R/hr/Ci at 1 meter.

With any detector you still need to know depending on the decay products the detection efficiency, make corrections for detector size, correct for radiation absorption due to intervening material between the source and detector ( especially important for alpha particles), correction for backscatter and in-source absorption.

The best way to determine activity is to choose a detector, source, geometry which minimizes the above problems.

Alphas usually have the advantage of being the most energetic particle in a decay. But there is another issue in using pulse mode ion chamber since the pulse heights due to an event depends on where in the chamber the event caused the ionization. The weaker the source the larger the volume must be and the more it exacerbates this problem.

The easiest method for half life determinations would be a pulse detection system sensitive to a key signature of the decay that you can detect with no or minimal interference from other radiation.

Isotopes like Cs 137 have one very energetic gamma along with a Beta particle that can be screen out.

WRT the dead time correction are you aware of the two source method for its determination?

One last question, did you have any particular isotope in mind?

 

[dRic2]

Typically ionization chambers are only used in current mode for relatively high radiation fluxes and for known sources. Ion chambers can be used to measure the activity based on the Specific Gamma Ray Dose Factor

As I suspected it seems to be easier to measure dose while operating in current mode, rather than activity.

One last question, did you have any particular isotope in mind?

No, I was just wondering. I am reading passages from Knoll's book and other notes on the internet and I keep running into examples where the detector is operated in pulse mode. It's fairly intuitive to link current pulses to counts per seconds. On the other and I'm not seeing a lot of activity measurements performed in current mode. Since current mode becomes necessary at very high activity I was wondering if there is a standard procedure to follow to measure activity in current mode. But, as I understood, current mode is best used when the source is know so that I can calibrate my detector before the measurement. If I had little information about the source I would definitely go for pulse mode since you can (in principle) obtain more information.

WRT the dead time correction are you aware of the two source method for its determination?

No, sorry. I have to look it up.

 

[bobob]

Move the source further from the detector so the rate is low enough to count particles and correct for the solid angle.