Answer to Question #12402 Submitted to "Ask the Experts"
Category: Instrumentation and Measurements — Surveys and Measurements (SM)
The following question was answered by an expert in the appropriate field:
When using a well counter for performing wipe tests, what is the highest background reading in disintegrations per minute (dpm) or counts per minute (cpm) that is acceptable?
Unfortunately, there is no single, well-defined quantitative answer to your question. This is because the acceptable background count rate depends on what you require for a minimum detectable activity (MDA), and the MDA depends on the background count rate. If the wipes are of radioactive sealed sources, the typical regulatory requirement in the United States is that the measurement system be able to detect at least as low as 185 Bq. Such a limit would not apply to contaminated surface area wipes, whose activity limits are generally expressed in units of activity per unit area. Since 100 cm2 is the typical recommended area to be covered by a single wipe, the results may often be reported in activity per area wiped, such as Bq per 100 cm2. The first requirement, then, in determining what maximum background count rate is appropriate is to specify what removable surface contamination level of the radionuclide(s) of concern is acceptable. You can then proceed to establish the necessary counting conditions to ensure sufficient analytical sensitivity.
When using a well counter for leak testing or for many other analytical applications, the usual procedure is to set up a region of interest (ROI) that includes the specific photopeak region associated with a gamma ray energy of interest. Naturally, more than one ROI may be established if more than one energy is of interest. One could argue that if we are concerned with only a single radionuclide, we could just as well collect counts over the whole energy range of the analyzer system. However, defining a region of interest does a couple of notable things: (1) it allows separation of the photon energy of interest from energies that might be associated with other radionuclides, and (2) it greatly narrows the energy range from what would be the case if one were simply to look at the counts over the whole energy range being covered and produces a great decrease in the background that will be detected. This can be to our advantage when measuring fairly low activities since, as said, the minimum detectable activity generally depends on the background count rate, as we show below.
If Rb is the background count rate in the ROI when no source is present a commonly used expression for the MDA, when equal rates of false positive and false negative results are acceptable, is
MDA = [k2/Ts + 2k(Rb/Tb + Rb/Ts)0.5 ]/E
where k has a value of 1.645 when the accepted rate of false positive and false negative results is 0.05, the most commonly used value in most health physics analyses; Tb is the time for which the background was counted; Ts is the time for which the sample was counted, and E is the counting efficiency for the radionuclide of interest, based on counts accumulated in the ROI in counts per disintegration. (Health Physics Society ATE Q/A 10110 discusses some of this.)
For example, a relatively small NaI(Tl) well detector, say 3.8 cm x 3.8 cm, might yield 30 to 40 cpm background in a ROI for 137Cs photons of 662 keV, while the background over the whole energy range from near zero to perhaps 1.5 MeV might be many hundreds of cpm, assuming a modest amount of shielding, say 1 to 2 cm of lead. If we assumed 40 cpm was the ROI background count rate, Tb was 10 minutes, and Ts was 1 minute, we would calculate the MDA for 137Cs as
MDA = [1.6452/1 + 2(1.645)(40/1 + 40/10)0.5]/E = 24.5 cpm/E.
For an assumed efficiency of 0.20 net counts in the ROI per disintegration of 137Cs (for wipes, this efficiency would apply to the geometry used, commonly the wipe folded up, often pushed to the bottom of a thin-walled tube or vial and the tube or vial placed into the well), the MDA would be 123 dpm, or 2.04 Bq, for 137Cs, based on the above.
If more than one radionuclide is present in the wipe, the background in the ROI may be different from that assumed above when no source is present if the second radionuclide produces counts in the desired ROI (e.g., Compton scattered photons from photon energies higher than that for the ROI). We shall not be discussing this further.
The bottom line is that you will have to input your own parameters, specific to your facility. The activity measurement requirement may be a regulatory one, such as that for sealed source leak testing or possibly one imposed as a license restriction, or it may be an in-house administrative limit. Once you have established this, you can then proceed to determine the requirements on your counting system and whether it is adequate to meet your needs. If not, there are measures that can be taken to improve the system’s performance. Such actions as increasing counting time, increasing shielding thickness, and increasing detector size are possible ways to improve detection limits. Because well counters typically have reasonably high efficiencies, even at relatively high gamma energies, I would not invest in a larger detector until I had reviewed how much greater efficiencies I would expect at the energies of interest for the large detector compared to the smaller one and whether such increases would meet your sensitivity needs.
I hope this adequately addresses your concerns.
George Chabot, PhD, CHP