|
13. Calibration of NaI(Tl) Detectors, Data Acquisition and Evaluation In Chapter 11>, the “energy versus channel” calibration of a NaI(Tl) detector system was discussed. Now, consideration will be given to the calibration of efficiency (counts per minute per nanocurie) versus energy. The NaI(Tl) calibration procedures are essentially the same for all Helgeson Scientific Services instruments:
This chapter covers the data acquisition and reduction of the data. We are assuming that the phantom for in vivo work is the Helgeson Scientific Services Simulated Livermore Phantom which was described in Chapter . The phantoms for waste monitors are the styrofoam filled barrels and/or barrels with source supports for filling with water, sand, etc. for higher density wastes. Calibrations are performed at a nominal 5, 6, 10, or 12 keV per channel using the HSS SLP Masonite phantom that has approximately standard man characteristics. A photograph of the phantom is found in Chapter 12>, Figure 12-9, and a drawing of the phantom is shown on page 13-3 of this chapter, Figure 13-1. Radionuclides are placed uniformly in the phantom in the same manner as they are expected to be found in the human body. Natural potassium and the cesiums are distributed uniformly throughout the phantom because they are materials that are distributed evenly throughout the total body. The cobalts and other non-transportable nuclides are distributed evenly in the lung cavity because the lung is the critical organ for these nuclides. 131-iodine is placed in the thyroid of the phantom because the thyroid is the critical organ for 131-iodine. Please refer to Chapter 12 for further details on source fabrication and distribution. The phantom is placed in the same counting position as a human (or waste barrel) would occupy during normal counting. Calibration data are accumulated for a sufficient amount of time to minimize the counting error of the calibration. This is a function of source strength. With approximately one microcurie of gamma radioactivity (this assumes that the necessary calculations have been made to account for gamma yield), the total counts should yield a small error, such as a few percent. For example, if 10,000 counts are obtained under the cesium-137 photopeak in 10 minutes, then the counting error (at one standard deviation) is + 100 counts, which is equivalent to 1% relative error. This also means that the counting statistics in the Compton scattered region will be reasonably good. The quality of the final results for counting people is a direct function of the quality of the calibration effort. Therefore, we recommend that the calibration counting times be as long as practical for best results. See Chapter 3 for statistical details. The “Do-It-Yourself” Lay Down Diagnostic Counter, the “Classic Shadow Shield Whole Body Counter,” and the “Quicky” In Vivo Counters are normally calibrated with one phantom size, the standard man. If a subject leans against the NaI(Tl) detectors of a “Quicky,” the results will be overestimated due to a change in counting position. The reverse is true if a person leans against the back of the counter. Since four detectors are used covering an effective distance of 112 centimeter (44 inches), the errors due to a person's position are smaller than they would be if a single 8 inch diameter detector were used in a fixed detector-subject relationship. It is recognized that errors are present when counting people who differ in stature from the standard man. This has been adequately demonstrated by the calibration work performed with the “Do-It-Yourself Whole Body Counter.” However, at this point, it is believed that these errors are acceptable since one of the purposes of the “Quicky” Counter is to sort people into two categories: those above and below the action point while still providing a reasonably good estimate of the body burden. Also, since the average counting time is 1 to 2 minutes, it is expected that people will stand erect during this time. Therefore, positional errors will be fewer. Complete instructions for obtaining calibration data for all systems are given in Chapter 14>.
13.1. Data Collection for NaI(Tl) Efficiency Calibrations
We are now ready to perform a sodium iodide efficiency calibration. The Energy Calibration program is reached by selecting the “Calibration” option from the Main Menu, see Figure 13-2, above, which shows the main menu. Please remember that this menu is the same for all “HELGE” software. Enter the program by selecting the “Calibration” option from the Main Menu, shown in Figure 13-2, above. To do this:
This action will provide you with one of the “Calibration” menus shown on the next two pages. Figure 13-3, at the top of the next page, shows the opening “Calibration” screen for a “Quicky I” while Figure 13-4, at the bottom of the next page, shows the opening “Calibration” screen for a “BRC” Waste Analyzer.
We are interested in “Efficiency” for all HSS instruments therefore:
If your System Manager has required a password for Efficiency calibrations, you will receive the screen shown in Figure 13-5, above. After providing the correct password you will see the “Data Entry” screen shown below in Figure 13-6.
The next screen depends on the type of system being calibrated. If you are calibrating an in vivo counter, you will see the screen shown in Figure 13-7, above, but if you are calibrating a waste monitor, look at Figure 13-8, below. The differences between the two screens are in lines 6 and 9 where we find Messages 728 and 731, respectively. We (Helgeson Scientific Services) normally change these messages for the appropriate counter at shipping time of the system. If question 2, the nuclide name, is answered exactly according to the abbreviations found in Table 13-1 found on page 13-13, then much of the information needed for the generation ICRP-30 or new Part 20 data will automatically be entered into the calibration column vector of counts per second per unit of activity (such as becquerels or nanoCuries).
Please refer to the types of questions asked. Most of these are self-explanatory, but we would like to emphasize several questions. 1. “Type of Calibration” is usually “EFFICIENCY.” 2. “Nuclide” should be answered, if possible, with the abbreviation found in Table 13-1, page 13-13 because this will speed up and simplify the evaluation process. 3. “Source I.D. Number” is the number that is found on the Source Calibration Certificate that is unique and identifies the source. 4. “Moving or Stationary” is self-explanatory. “Quicky” counters are always “Stationary” (expect “Quicky IV” and “Quicky V”), the “Do-It-Yourself” Lay Down Diagnostic Counter may be either, and the Waste Analyzers are usually “Moving.” 5. “Source Distribution” is usually “UNIFORM,” but should state the organ, such as “UNIFORM LUNG,” or “UNIFORM TOTAL BODY. 6. “Number of Spacers” (for the in vivo counters) means the number of extra Masonite spacers used to increase the height of the phantom, while “Number of Rods” (for the Waste Monitors) means the number of rod sources used in the styrofoam phantom. 7. “Units” is self-explanatory. 8. “Activity reference date” is the date obtained from the Calibration Certificate on which the source strength was determined.
9.
“Source activity per dot” (for the in vivo counters) is self-explanatory
- what is the reference activity of one “dot” on the activity reference
date? This is obtained by dividing the total source activity, found on
the Calibration Certificate, by the number of “dots.” 10 “Count Time (sec)” is self-explanatory - how long in seconds do you want to count the phantom?
Figure 13-9, above, shows the screen for a BRC Waste Analyzer when the rotating drum support is not at its normal starting position. There is nothing that the operator need do other than wait for the software to automatically move the drum support to its starting position. This movement is always done at high speed.
Figure 13-10, below, shows an Efficiency Calibration spectrum when the data collection is about 75% done (look at the Live Time versus the Real Time).
Figure 13-11, above, shows the end of a count. The data from each of the detectors and the composite spectrum (the sum of all the detectors) are written to the WB1: directory. The data are now ready for conversion to counts per second per unit of activity, typically becquerels.
13.2. Evaluation of the Calibration Data Now that we have collected the calibration data it must be evaluated, or reduced, to a form which may be used by the standard Least Squares Analysis software which was discussed in Chapters , “Data Acquisition and Analysis,” , “Data Analysis,” and , “Least Square Theory.” This form is the “single column vector,” a mathematical term, and is more easily called a “calibration column” because it is a single column of calibration values, channel-by-channel, which have units of counts per second per Becquerel. For users of the “BRC” Waste Analyzer you may also generate nuclide specific “CAL” files. This will be discussed further starting on page 13-61.
Figure 13-12, below, shows the “calibration” menu selection which you must use to reduce the data to a usable form:
The program may be password protected. If so, you will obtain the screen found on the next page. Option 4 allows you to generate the single column vector calibration file which is used in the Data Analysis program. This file has units of “counts per second per Becquerel.” Although the type of instrument shown in Figure 13-12 is the “BRC,” this section, “Evaluation of Calibration Data,” applies to all HSS NaI(Tl) instruments. Option 5 is used only for the “BRC” counter and will be discussed in Section 13-3 of this chapter starting on page 13-61. As we saw in Figure 13-3, page 13-5, Option 5 does not exist for in vivo counters. But before we start on data reduction we must look at the “ELEMENTS.INF” file and discuss its importance.
13.2.1. The “ELEMENTS.INF” File As we learned in Section 13.1, “Data Collection for NaI(Tl) Efficiency Calibrations,” you must enter the data in a particular format. If you have, the software will ask only a few questions:
If you do not, then the software will ask additional questions, such as:
These last data are contained in the “ELEMENTS.INF” file. If the questions of Section 13.1 have been answered exactly, then the calibration evaluation program will obtain its needed data directly from the raw data file. Table 13-1, found on the adjacent page, is a copy of the “ELEMENTS.INF” file. It contains the abbreviation of the nuclide, its full name, the energy in keV of the principal peak, the center channel of the peak, the half-life of the radionuclide, and the units of the half life. The meanings of the half-life units should be obvious: M = minutes, H = hours, D = days, Y = years. In this section we are defining “Common Nuclide” to mean radionuclides which are listed in the “ELEMENTS.INF” file. Thus, by the above definition, any nuclide which is not in the “ELEMENTS.INF” file is considered an “Uncommon Nuclide.” This file may be edited with any ASCII text editor to insert additional nuclides so that what may be an “Uncommon Nuclide” at a nuclear power station, for example, 238-uranium which is used in the section of “Uncommon Nuclides,” may become a common nuclide at a facility which works primarily with 238-uranium.
Frame 32
13.2.2. Evaluation of a Nuclide Found in the ELEMENTS.INF File Evaluating efficiency calibration data involves several steps:
13.2.2.1 Start the program, Choose and Display the file We are now ready to reduce the raw data to a calibration column vector. Enter the proper password, as requested in Figure 13-13 at the top of the next page, and the program proceeds to the information screen shown at the bottom of the next page in Figure 13-14. This screen describes the general steps which will be taken in the evaluation of the calibration data file in the process of reducing it to the single column vector which is used in the Least Squares program. Normally, you will respond “Y(es)” to the question. We are now ready to select a file for evaluation.
The program immediately provides the screen shown in Figure 13-15, below. Note the upper left corner of the figure where you will see “WB1\*.cdn” displayed. This means that you are looking at only the files with the extension “.CDN.” (The program accepts either upper or lower case extensions.) You may change to a different extension by pressing “F3.” We won't do that in this example, rather, we shall show this to you in the section on “Uncommon Nuclides.” See Figures 13-42 through 13-45 if you want information on this subject at this time. Figure 13-15, below, shows that file P3191529.CDT has been highlighted. We will select it by simply pressing the <ENTER¿> key. We have selected file P3191529.CDT from the list shown in Figure 13-15. It is displayed in Figure 13-17. Look at the center of the screen shown in Figure 13-16 at the top of the opposite page. The program has asked: Smooth the spectra? (Y/n) Smoothing will minimize statistical errors. This subject has been discussed in detail starting on page - of Chapter . Typically, you should answer “Y” to this question. You do not need to press <ENTER¿>, the program will continue as soon as you press the “Y” or “N” key. The smoothed spectra are now displayed in Figure 13-17. Note that these is a “message box” at the bottom of the upper left quadrant. Keep an eye on this box as it will provide messages telling you what action should be taken during the calibration evaluation.
13.2.2.2. Mark the Low and High Channels for Gain and Zero Shift Our next task is to perform a gain and zero shift to bring the data to standard reference conditions. Note that the message box in the upper left says: Mark the low energy peak with PgUp. Do this in the normal manner by moving the variable cursor to the left (it is now in channel 256). Remember, pulsing the left arrow key moves it one channel at a time while holding the “CTRL” key and pressing the left (or right) arrow key moves the cursor 32 channels at a time. As you move the cursor the counts per second are displayed in the upper right hand corner of the spectral window. Looking at these values will help you find maxima and minima values. In this example we are marking channel 14 which is the center of the 241-americium photopeak. Figure 13-18, found at the top of the next page, shows the result of marking channel 14. Note that the message box in the upper left corner has now changed to Mark the high energy peak with PgUp. Obviously, you do this in the same way. Figure 13-19, found at the bottom of the next page, shows that we have marked the lower 60-cobalt photopeak. It has an energy of 1173.21 keV. Note that the message box area of Figure 13-19 now says “Press Alt-G for Gain/Zero Shift.” When this is done you will see a pop-up window on the right side of the upper left window area. Please turn to page 13-20 and refer to Figure 13-20 for the use of this pop-up window.
13.2.2.3. Choose the Energies of the Marked Peaks
When you press “Alt-G,” you will see a pop-up menu on the right side of the upper left window area with a list of nuclides. Look at Figure 13-20, below. This list comes from the “ELEMENTS.INF” file. The message at the top of the upper left window is asking: Low energy nuclide: This list starts with four question marks. If the photopeak which you have chosen is not in the ELEMENTS.INF list, you may enter appropriate values. We have chosen to demonstrate this in Figure 13-21 shown at the top of the next page. When you press <ENTER¿>, the three questions are asked one at a time. Respond to each of them with <ENTER¿>. When the last question has been answered, the program asks: High energy nuclide. But we are getting ahead of ourselves. In Figure 13-22 we have moved the highlighted region to the 241-americium box. We will select the photopeak which we chose previously, 241-americium, by pressing <ENTER¿>. Now please turn to page 13-22.
The result is shown in Figure 13-23, above. The “ELEMENTS.INF file has been accessed and the 241-americium data have been retrieved and displayed. The message at the top of the upper left window is asking: High energy nuclide:
and the highlight is in the four question mark box. In Figure 13-24, below, we have moved the highlight to “CO60”. Now press <ENTER¿>.
The result is shown in Figure 13-25. The program message: Is everything OK? (Y/N) cautions you to be sure to check the results to see if they are correct. If they are not, respond “N” and the program will return to the beginning of the energy selection process, page 13-20.
13.2.2.4. Perform the Gain and Zero Shift
In the previous section we have marked the low and the high channels preparatory to actually shifting the gain and zero to standard reference conditions. When you press <ENTER¿> the program will print the results of the gain and zero shift as shown in Figure 13-26, below, and with this the gain/zero shift portion of the program is finished. We are now ready to select the region of interest which is used to calculate the minimum sensitivity.Follow the program instruction: Press any key to continue...
13.2.2.5. Define the Region of Interest (ROI)
We start the process of defining the Region of Interest (ROI) with Figure 13-27, below. This process involves marking the low and high channels of the ROI by moving the variable cursor to the chosen channel, then pressing the “PgUp” key. The sum of the counts per second per unit of radioactivity (Becquerels, for example) in the ROI is used in the calculation of the minimum sensitivity. The minimum sensitivity is used to determine whether or not a calculated result is to be kept as a part of the results. If the result is less than the “Minimum Sensitivity Drop Factor,” then the nuclide is dropped from further analysis. Please refer to Chapter , page - for more information on the “Drop Factors.” When we enter the ROI portion of the program, the screen shows Mark Start ROI with PgUp. Initially the variable marker is in channel 256 as shown in Figure 13-27. Move the variable marker to the beginning of the ROI and press the “PgUp” key. In Figure 13-28 on page 13-26 we have moved the variable cursor to channel 214, which is on the low energy side of the 60-cobalt 1,173.2 keV photopeak.
Figure 13-28 shows the lower marker selected. The message at the bottom of the upper left window says Mark end of ROI with PgUp.
Move the variable marker to the high energy end of the ROI. Looking at the upper right window of Figure 13-28 you can see that the variable marker has been moved to channel 245. Now press the “PgUp” key. When you do, two messages are displayed:
Store zeros in channel 214 and in the message box: Alt Z to zero channel between markers See Figure 13-29 at the top of the next page. The message in the upper left window shows that the ROI is where we have requested it to be. However, you must be careful at this point because the current version of the software needs some improvement. Normally, you would expect that a marker would be erected in the selected channel, 245, but instead, the first marker is erased, the end ROI is stored in memory and the program proceeds to the instruction: Alt Z to zero channel between markers (Editor's note: If you watch very carefully you will see that the second marker is erected, but the process is so fast that it appears that the first marker is erased.) Obviously, we don't want to place zeros in our region of interest, so respond “N.” This will take us to the part of the program where we can really store zeros in certain channels. Please turn to page 13-28 for the next part.
13.2.2.6. Store Zeros in Selected Channels A 241-americium source is almost always placed adjacent to the NaI(Tl) detector(s) in the Helgeson in vivo counters. We know the energy of the principal photopeak, 59.537 keV, and since the source(s) are permanently fixed with relationship to the detector(s), the constant geometry allows us to obtain a measure of the collection efficiency of the NaI(Tl) detector. The integral of the counts per second beneath the 241-americium photopeak in the background and the center of the photopeak are stored as Quality Control values 7 and 1 (before gain & zero shift) and 5 (after). However, because of the random statistical fluctuations of counts in the low energy region due to the subtraction of a background containing the 241-americium photopeak from a calibration count, also containing the 241-americium photopeak (typically channels 1 through 20 or energies of 5 through 100 keV), these 20 channels are of little value in a calibration column vector. Therefore, we normally set these channels to zero counts per second per unit of activity. Additionally, we will also see random statistical fluctuations of counts in the energy regions above the highest energy of all radionuclides. Thus, we want to set these to zero because they also have no value in the Least Squares analysis process. This portion of the program will ask you if you wish to store zeros in two groups of selected channels. There are only two energy possible regions, one at the low energy end of the spectrum and the other at the high energy end. This portion of the program starts by asking: Do you want to store zeros in a group of channels? (Y/N) Frame 80Answer “Y” to proceed. If you answer “N” the program goes directly to the last part where you name the file, enter the ALI, etc.
The program asks for the first channel of the group in Figure 13-32: Mark the low channel with PgUp 
Now the program asks for the last channel of the group in Figure 13-31: Mark the high channel with PgUp You can see the two markers in Figure 13-31 between channels 1 and 17.
In Figure 13-33 the program is asking: Is everything OK? (Y/N)
Respond “Y” if true and the program will zero the data in these channels and go back to the beginning of this part to see if there are other areas to be zeroed. Figure 13-34 shows the resulting display when the channels are zeroed. Respond “N” if you want to change the channels for this region. 13.2.2.7. The Remaining Information Frame 92The last part of the generation of a single column vector of calibration information requires answers to the following questions: File name for saving: Enter a file name which is appropriate to the nuclide and with an extension which is consistent with other extensions in the \SUPPORTS sub-directory of calibration data. ALI: This is the Annual Limit of Intake which is printed in ICRP Publication 30, Supplement to Part 1 (Reference ICRP30(S1) found in Chapter ). If you are using the ALI values from the revised Part 20 (Reference FR21May91), then use those values. Print Results? This is the same question asked in the Parameters portion of the program. Do you want the results for this nuclide printed on the output page? Action point: Enter the action point in consistent units, Bq, nCi, etc. Nuclide Name: Enter the nuclide name as you wish it to be printed on the output page. Your Initials: Enter the initials of the technician who performed the calibration analysis. Figure 13-36 at the top of the next page shows typical answers to these questions.
You can use Norton Commander or some type of ASCII text editor to see the output form of the file which we have just created. Figure 13-37, below, shows the \SUPPORTS sub-directory. Note that the file which we generated is listed there. By using the Norton Commander “F3” key we may examine this file. Please turn the page to Figures 13-38 and 13-39 to see the beginning and the end of this file.
Figure 13-38 shows the top of the file. Note that name of the nuclide, “CO60,” requested at the time the calibration data were acquired (see Figure 13-7, page 13-7) has been repeated here without being entered again. Also shown are the initials of the technician and the source strength at the time the data were accumulated (compare 37,000 Bq from Figure 13-8, page 13-7 with 22,207.22 Bq in Figure 13-38, above).
Figure 13-39, above, shows the end of the file. Note that the ROI, ALI, Print request, Action point and Nuclide name have all been copied. These are used in the analysis program. You may modify these values after the file has been generated by using the capabilities in Chapter , “Examine/Change Parameters.” Each time you modify one of these values in the Parameters chapter, the changes are written to the corresponding single column vector calibration file. Tables 13-40 and 13-41 are found on the next two pages. These tables are really only one table and present the entire calibration column vector with one exception: all of the data found at the end of the file and shown above have not been repeated, although they are truly in the file. We have now shown you how to generate a calibration column vector when you have used the same abbreviation for the nuclide as was found in the “ELEMENTS.INF” file, Table 13-1 found on page 13-13. Now let us examine the method of evaluating a file whose abbreviation is not in the “ELEMENTS.INF” file. Please turn to page 13-38. Frame 105
Frame 107
13.2.3. Evaluation of an Uncommon Nuclide Now let us consider a calibration file for which there are no data in the “ELEMENTS.INF” file. This part of the program is very similar to the section on nuclides which are found in the ELEMENTS.INF file. 13.2.3.1 Choose and Display the File
We shall not repeat all of the previous menu graphs, rather, we pick up the process just after the information screen, Figure 13-42 at a point equivalent to Figure 13-15, page 13-16. In this example we shall use 238-uranium and determine the calibration column vector for the top detector in a “BRC” Waste Analyzer. We could have inserted the 238-uranium data into the file, but then we would have had to choose a different nuclide for our example. Note that the file filter shown in Figure 13-42 is “WB1\*.cdn.” We want to look at all of the files. To do this we must press the “F3” key as shown in Figure 13-43 on the opposite page.
The F3 key gives you the screen shown in Figure 13-43, above. (Editor's Note: This is not exactly correct: the down arrow has been pressed to select “all files,” the “*.*” filter. It is always on the top row when you enter this screen). Having selected the “*.*” filter, the program asks you to wait while it reads the “WB1” directory and compiles the list of files, as seen in Figure 13-44, below.
On completion of the search, Figure 13-45 is presented. You may now move the up- and down-arrows to select the file which you wish to evaluate. File A3200945.1DT has been chosen. Note the “E” at the far right of the list. This means that this is an “Efficiency” file as contrasted to those files which do not show the “E.” We reached the screen shown in Figure 13-46 at the bottom of the next page because the nuclide name was not been found in the file “ELEMENTS.INF.” You have a choice:
In this example we shall continue with the analysis by responding “Y.” You do not need to press <ENTER¿>, the program will continue as soon as you press the “Y” or “N” key. The balance of this example uses the 238-uranium data file which was chosen in Figure 13-45, page 13-40. Before we continue, let us look at the upper right portion of the screen, the portion headed with the term “Helgeson BRC.” The mode shows “Make Calb File.” This means that we are in a program called “MAKECALB.EXE.” The next line shows that the length of the count is 0 seconds. Since this is not a data acquisition, we are using the same screen but we do not need to show anything for the length of the count. The channel number shows where the variable marker is located. The next line shows the number of counts in the variable marker. We shall see later that this entry contains a real number. The next two lines are self-explanatory. The last line states “CTRL F1 - Exit MAKE CALIB.” This is merely an instruction that if you wish to leave the program before you have finished you can exit at any time by down the CTRL key and pressing F1. This action will return you to the calibration menu, Figure 13-12.
When you have answered the “Continue ?” question with “Y,” the program will ask if you want to smooth the data. Look at the center of the screen shown in Figure 13-47, below. Smoothing will minimize statistical errors. This subject has been discussed in detail starting on page -. You do not need to press <ENTER¿>, the program will continue as soon as you press the “Y” or “N” key.
13.2.3.2. Identify the Nuclide
This step is one which differs from the evaluations in which the nuclide has been properly identified in accordance with the protocol for the file ELEMENTS.INF. Respond to the question “Nuclide abbreviation,” as shown in Figure 13-48, below. The “Nuclide abbreviation” is typically entered without a hyphen, but since anything entered in this portion will not be entered into the “ELEMENTS.INF” file, it doesn't make much difference. When you have entered the abbreviation press <ENTER¿>. Figure 13-49, at the top of the next page, shows that we have entered the nuclide abbreviation, the half-life, and the half-life units. When you have entered the half-life units, press <ENTER¿> and the program continues, see Figure 13-50 at the bottom of the next page, by asking for the source strength (always in Becquerels) and the date on which the source strength was determined. This information should be on the source calibration certificate.
13.2.3.3. Mark the Low and High Channels for Gain and Zero Shift This section is almost identical to that found on page 13-18. The same information has been repeated here so you will not have to turn back to see what is to be done. Our next task is to perform a gain and zero shift to bring the data to standard reference conditions. Note that the message box in the upper left says Mark the low energy peak with PgUp. Do this in the normal manner by moving the variable cursor to the left (it is now in channel 256).The variable marker is typically in channel 256 when you first reach this screen. Remember, pulsing the left arrow key moves it one channel at a time while holding the “Ctrl” key and pressing the left (or right) arrow key moves the cursor 32 channels at a time. As you move the cursor the counts per second are displayed in the upper right hand corner of the spectral window. Looking at these values will help you find maxima and minima values. In this example we are marking channel 40 which is the center of the 235-uranium photopeak at 185.70 keV. Before you press “PgUp,” move it a few channels left or right, looking at the upper right hand corner of the graph area. Note in this case that in channel 40 (look at the box directly under the “F4-Info” marker) we see the subject and background counting rates in counts per second. In this particular example we had a maximum subject counting rate of 0.2299 counts per second when the variable marker was in channel 40. Now you should press “PgUp.” The result is shown in Figure 13-51 found at the top of the next page. Note that you have received a red marker and that at the top of the graph area the number 1 is shown indicating that this is marker number 1. The message box in the upper left corner has now changed to Mark the high energy peak with PgUp. Obviously, you do this in the same way. In Figure 13-51 we have already moved the variable marker to the center of the 40-potassium photopeak which we found in channel 282 with a maximum counting rate of 0.0241 counts per second. Now we may press “PgUp” and this gives us the second marker as shown in Figure 13-52. Once we have marked these two photopeaks we are ready to obtain information for the Gain/Zero shift. The message window says “Press Alt-G for Gain/Zero shift.” Note that the message box area of Figure 13-52 now says “Press Alt-G for Gain/Zero Shift.” When this is done you will see a pop-up window on the right side of the upper left window area. Please turn to page 13-20 and refer to Figure 13-20 for the use of this pop-up window.
13.2.3.4. Perform the Gain and Zero Shift
We enter this part of the program with the highlighted box in the “????” area as seen in Figure 13-53. Since we have already marked the low and high channels, we must now find the energies of these two photopeaks. Note in the upper left hand corner that the program is asking for the name of the low energy nuclide. Since we have chosen 235-uranium, use the down arrow to find the nuclide. In Figure 13-54 we have moved the highlighted region to the 235-uranium box. We will select the photopeak which we chose previously, 235-uranium, by pressing <ENTER¿>. The result is shown in Figure 13-55. These data have been taken from the ELEMENTS.INF file which we discussed previously. Note that the center peak for 235-uranium is expected to be in channel 39.06 and the energy is 185.70 keV. The program now asks for the high energy nuclide. In Figure 13-54 we have found “U-235.” When we press the <ENTER¿> key we obtain the screen shown in Figure 13-55. Note that the “????” area has been highlighted again and that the programs is asking: High energy nuclide.
In Figure 13-56 we have moved the highlighted area to the “K40” area. Figure 13-57 at the top of the next page shows the results obtained when we have pressed the <ENTER¿> key. The program picked up the center channel of 284.52 with an energy of 1460.73 keV from the ELEMENTS.INF file. At the bottom of the upper left window of Figure 13-57 found at the top of the next page the program is asking Is everything above OK? (Y/N) If you are satisfied with these entries press “Y” without pressing the <ENTER¿> and the program will continue. If you are not satisfied press “N” and the program will return to where we were in Figure 13-53. Assuming that everything is satisfactory, the program finds the center channels of the two photopeaks, determines their resolution, and does the gain and zero shift. Figure 13-58 shows the results of these calculations. The Gain and Zero Shift operation has now been completed. To continue the program press any key. Please turn to page 13-50.
13.2.3.5. Define the Region of Interest (ROI)
Figure 13-59, below, shows that the program is now asking for us to: Mark Start ROI with PgUp In this particular example we are using this calibration file for a “BRC” counter. A BRC Waste Analyzer is designed to count everything from essentially 50 keV to well over 2 MeV. An in vivo counter typically has a 241-americium source near the detector to act as a known low energy reference point. Therefore, the calibration column vector usually contains zero in channels 1 through 20. In this example we want to start our ROI in channel 11. In Figure 13-59 you can see that the variable marker has been moved to channel 11. Note the position of the variable marker. Normally, the variable marker is above the spectrum. In this figure the variable marker is below spectrum because it cannot exceed the upper boundary of the spectrum screen. The channel number is shown in the upper right hand corner. When you have found the channel you wish, press the “PgUp” key. This will produce Figure 13-60 which shows that the starting region of interest is channel 11. Channel 11 was chosen because we do not wish to work with energies lower than that in this example. In fact, we will be deleting all the counts from channel 1 through channel 10 at a later point. Figure 13-61 shows that we have moved the variable marker to channel 500. (Look in the right hand box for the channel number and look at the graph at the extreme right hand side where you will find the variable marker.) We are now ready to press PgUp for the high energy marker.
Figure 13-62, above, show the result of pushing the PgUp after we have marked the high energy marker. Notice at the bottom of the upper left box that we now have the instruction: Alt Z to zero channel between markers.
We shall wish to eliminate random statistical fluctuations which have an average value of zero. However, due to a software omission we have a slight problem at this point. When we press “Alt-Z,” we obtain the picture shown in Figure 13-63, below. We certainly don't wish to store zeros in our region of interest, so respond to this question with “N.” This will bring us to the real portion of the program for storing zeros in selected channels. Please turn the page. 13.2.3.6. Store Zeros in Selected ChannelsThis portion of the program will ask you if you wish to store zeros in two groups of selected channels. There are only two energy possible regions, one at the low energy end of the spectrum and the other at the high energy end. This portion of the program starts by asking: Do you want to store zeros in a group of channels? (Y/N) See Figure 13-64, below. Answer “Y” to proceed. If you answer “N” the program goes directly to the last part of the program where you name the file, enter the ALI, etc. In Figure 13-65, at the top of the next page, we have moved the variable cursor to channel 1 because we are going to zero channels 1 through 10. Figure 13-66, at the bottom of the next page, shows that we have moved the variable marker to channel 11.
When we push “PgUp,” we obtain Figure 13-67, above, with the question: Store zeros in channels 1 to channel 11 ? 
Since our objective was to store zeros in these channels, we shall respond with “Y.” Figure 13-68, below, shows that we have accomplished our goal. We have also moved the variable marker to channel 501, preparatory to zeroing channels 501 through 511.
In Figure 13-69, above, we have pressed “PgUp” and we have a marker in channel 501. In Figure 13-70, below, we have moved the variable marker to channel 511.
In Figure 13-71, above, we have pressed “PgUp” and we now have markers in channels 501 and 511. When we press “Alt-Z,” the computer asks: Store zeros in channel 501 to channel 511 ?
Since our objective was to store zeros in these channels, we shall respond with “Y.” Figure 13-72, below, shows that we have accomplished our goal.
Figure 13-73, above, shows that we have accomplished our goal. We are now ready to move on to the last portion of the calibration work.
13.2.3.7. The Remaining Information This part is essentially identical to that found on page 13-30. The last part of the generation of a single column vector of calibration information requires answers to the following questions: File name for saving:
Enter a file name which is appropriate to the nuclide and with an extension which is consistent with other extensions in the \SUPPORTS sub-directory of calibration data. See Figure 13-74, below, for an example of the software problem. ALI: This is the Annual Limit of Intake which is printed in ICRP Publication 30, Supplement to Part 1 (Reference ICRP30(S1) found in Chapter ). If you are using the ALI values from the revised Part 20 (Reference FR21May91), then use those values. Print Results? This is the same question asked in the Parameters portion of the program. Do you want the results for this nuclide printed on the output page? Action point: Enter the action point in consistent units, Bq, nCi, etc. Nuclide Name: Enter the nuclide name as you wish it to be printed on the output page. Your Initials: Enter the initials of the technician who performed the calibration analysis. Figure 13-75 at the top of the next page shows typical answers to these questions.
Figure 13-76, below, shows the six questions which are asked before the data are written to the disk. First of all, we need to have a name for the file. In this example we gave it the name “U238X.BRC.”
13.3. Make a “BRC” Calibration File A typical efficiency curve for the NaI(Tl) detectors as a function of energy is shown on the next page. Presented in this graph is the response of the detector for activity deposited in the phantom lung. This page is for Figure 2 - a typical efficiency curve for the NaI(Tl) detectors as a function of energy. Figure 2 shows the response of the “Quicky” Counter to Barium-133 (used as “mock-iodine”) which has been placed in the “thyroid” of the phantom as a function of the height of the phantom. Since the phantom is designed to represent a typical human being, this curve shows the change in calibration factor as the thyroid is raised above the center line of the No. 1 (top) detector. The height of the person is asked as an input parameter to this function. The computer obtains the height and thereby calculates the correct calibration factor for radioiodine. Once a series of files has been generated in units of net counts per second per becquerel, the data may then be placed in the calibration matrix. The mathematical algorithm used to determine the amount of radioactivity in a subject utilizes a modification, or simplification, of the classical “Least Squares.” This algorithm uses multiple regions of interest, providing an over-determined system of equations. Rather than perform the analysis on a channel-by-channel basis, as is the normal convention with the “Least Squares” process for gamma spectra, groups of channels are summed and treated as a datum point. This over-determined system of equations is then solved by normal methods: Matrix transform, matrix inversion, generation of the answers, along with the error terms. Thus, it becomes necessary to choose the regions of interest prior to starting the QUMATX program. 13.4. Compton Scatter Within in any shadow-shield type counter there is the problem of high energy background, cosmic ray, and natural terrestrial radiation backgrounds. These gamma rays will strike the person being counted and may scatter by pair production or Compton scattering process. Within the scattering process is the degradation of the primary energy of the gamma ray. The amount of scatter is a function of the energy of the incident radiation, the angle of incidence, the atomic number of the scattering material, and the mass of the scattering object. In any given location, the background radiation is normally fairly constant with respect to energy. (This is not necessarily true during venting of off-gasses at nuclear power sites.) Likewise, the angles of incidence are constant. The atomic number of the scattering object (the human body) is constant. Thus, the only factor that varies is the mass of the scattering body. Studies have shown that the amount of scatter is truly a function of the weight of the person, as would be predicted from theory. However, the slope of the scattering curve as a function of energy appears to be constant regardless of the scattering mass. If a scattering component is not included in the calibration matrix, then there will always be high results for iodine-131, or other low energy radionuclides that may be in the matrix. With the addition of a Compton scatter component, these incorrect high iodine results disappear. Since the slope of the Compton curve is constant for all masses, then one Compton component curve may be added to the matrix and the computer will solve for the fractional amount needed to lower the errors. This will make the iodine result more realistic. The Compton component is best determined by counting many people who have no known radioactivity other than naturally occurring potassium. The spectra may be obtained with the SPCTRA program along with an appropriate background, or may be obtained from the “Fixed Time Count,” Option 5 in the “'Quicky' Count” Menu. The potassium contribution is subtracted from the net spectrum using the SPCTRA program. The resulting net spectrum is stored for each person. After 20 or more people have been measured, their net spectra are recalled by the SPCTRA program, added together, and converted to a single curve. They may be normalized to “unity” by dividing by the number of people counted. This will give the average counts per second per channel for the average person. These data may then be incorporated into the calibration matrix.
This is the end of Chapter 13,
|
