CESIUM 2k ANALYZER – A PORTABLE INSTRUMENT FOR ¹³⁷Cs IDENTIFICATION AND COUNTING

Norman Latner, Norman Chiu, and Anna Berne

Environmental Measurements Laboratory

U. S. Department of Energy

New York, NY 10014-4811

Introduction

The Cesium 2k Analyzer (Figure 1) was designed by the Environmental Measurements Laboratory (EML) to evaluate the activity of ¹³⁷Cs collected on Empore Rad disks. The cesium specific Empore disks are an experimental product, and prior to the development of this unit, no portable instrument was available for evaluating the activity collected on the disks. Empore Rad disks, developed jointly by 3M Industrial Products and Argonne National Laboratory, are densely packed sampling disks fabricated with element selective particles to separate, collect and concentrate target radioisotopes on the surface of the disk. Because they exhibit a low liquid flow resistance, and are relatively immune to different elements that simply pass through the disk, a high recovery of the target isotope can be achieved. These Rad disks are ideal for quantitative determination of a specific isotope in aqueous solutions, offering an effective alternative to standard radiochemical sample preparation methods.

Figure 1. The Cesium 2k Analyzer.

Design

The EML Cesium 2k Analyzer identifies and counts the 662 keV gamma-rays that are emitted from the Empore Cesium Rad disks (Figure 2), subtracts out the previously collected background in proper proportion, and shows the net activity on its digital display. It is configured as a 32-pound portable field instrument that will operate for 56 h after recharging. The unit uses an on-board cesium-tracking source to establish a calibration before use at a location, thus allowing the instrument to operate over a wide range of environmental temperatures and conditions. While the design is based on a microcomputer, making it easy to operate, every effort was made to keep the cost below $2,000, rather than the $10-12,000 it typically costs using a more conventional multichannel analyzer and notebook computer approach.

This unit was designed for simplicity of operation and has only five operating keys. These operating keys allow the user to choose a counting time, select the calibration procedure, or start a background or sample run. The keys also permit a run to be aborted at any time. The two-line LCD display shows information that relates to, and identifies, the mode that the unit is in. While counting, a flashing display is shown, as well as preset and elapsed counting times. Background count is shown both as total count for the preset period, and in counts per minute (cpm). The net sample data is shown as a total count, but the program can be modified to indicate activity in radiation units. The calibration screen shows a tracking number that gives the user an indication of how much correction was supplied.

The design of this instrument often involved a balance, or compromise, between opposing elements. The $2,000 price constraint limited the choices to comparatively inexpensive detector, computer and circuit elements, and, thus, increased the design complexity. As a portable instrument, there was a need to minimize the shield to keep the weight down, but at the same time, more shielding lowers the background and leads to greater sensitivity. Similarly, a larger size detector increases sensitivity, but at the cost of a larger and heavier shield. Using a $49 computer (Basic Stamp II) was cost effective, but required much time-consuming programming. At times it was possible to use the same components in several ways to achieve economies. Rather than using a single 12 V lead acid battery, two 6 V series wired units were used. These batteries were arranged so as to straddle the detector, thus effectively augmenting the shielding, and further reducing the background by 18%.

Figure 2. Empore Cesium Rad Disks.

Operation

The first step in operating the Cs2k analyzer is loading the on-board calibration source and starting the automatic calibration routine (Figure 3). This procedure corrects for any gain shifts and establishes the proper starting point. The process may take 1 to 3 min while the computer compares the number of counts in two adjacent counting windows (Figure 4). The computer successively shifts both windows towards the higher count region, until the counts in both windows are equal and the cesium 662 keV peak is centered.

A sample blank is counted next to yield a background count. The preset time is set by the increment/decrement keys and adjusted to a value between 1 and 100 min. A longer background count is always advantageous when the expected sample count is low, since this reduces the standard deviation and increases the confidence of the result. Only one background count needs to be taken per location. This stored count will be proportionally applied to all subsequent sample counts.

Prior to counting a sample, the protruding luer fittings are clipped off both sides of the Empore Rad disk using the on-board clipper. A count time is then entered and counting started. Here again, a longer count leads to better statistics. At the completion of counting, the display shows preset time, elapsed time, total gross count and total net count.

Figure 3. Loading a sample into the unit.

Figure 4. Automatic calibration routine.

Brief description of the Cs2k analyzer circuitry

The block diagram is shown as Figure 5. At left is the detector, a 2x2-inch sodium iodide inline unit with the crystal surrounded by a 0.5-inch lead shield. The shield section that covers the end of the detector also serves as the sample holder. A commercial preamplifier and base unit was used. Bias for the detector is supplied via a d.c. to d.c. high voltage converter module whose output is adjusted by a precision regulator circuit.

The signal from the preamplifier, having an amplitude proportional to the detected gamma energy, is routed to three comparitors. Each comparitor is adjusted to a particular level and only pulses exceeding that level pass through that comparitor. As shown in the figure, level A includes the entire cesium peak and above, level B encompasses the middle of the peak and above, while level C accepts only the counts above the peak. The critical voltages that set the comparitors are actually generated by the “Basic Stamp” computer software and thus can be easily adjusted. This process is accomplished by converting the computer’s digital output to the analog levels needed for the comparitors, using Digital-to-Analog converters and a precision reference. The pulse outputs from the comparitors are accumulated in three counters and periodically shifted into the computer.

The computer, programmed in basic, forms the counting windows A-B and B-C, which are combined to count the entire cesium peak, or used separately during the calibration routine. During this procedure, a short run is made and the counts in A-B and B-C are compared. If they are not equal, levels A, B and C are simultaneously shifted in the direction of the higher count window and a new run is made. The process is repeated until the windows are balanced, at which point the unit is energy calibrated. The computer controls all display functions. A crystal clock circuit supplies the basic clock functions needed by the computer.

Figure 5. Block diagram.

Results and Discussion

The Cs2k analyzer exhibits a broad measurement range of from less than 50 pCi to over 400, 000 pCi. At the higher counting rates, short count times are adequate and the background contribution becomes less important. A series of replicate counter measurements were made, using different source strength samples, to give statistical assurance of counter reliability. The background rate during these tests was about 70 cpm. For a source rated at 141,800 pCi, and a count time of 5 min, the mean count was 20,135 " 80 cpm. Since the calculated value of 2 sigma (95%) is " 282 cpm, these data are quite good. A lower activity source of 15,080 pCi counted for 10 min gave a mean value of 1864 " 18 cpm, compared to " 86, the calculated value of 2 sigma. At levels approaching the lower limit, a 60 pCi sample, using 60 min count times and a 60 min background, results in a mean value of 6.2 " 2 cpm. These results are good since the calculated 2 sigma value is " 3 cpm. Finally, 20 min runs with a

120 pCi sample and a 60 min background gave mean values of 13.3 " 4.0 cpm compared to the calculated 2 sigma value of " 4.6 cpm.

The Cs2k analyzer was developed for rugged field use and was built using a heavy-duty aluminum case. With 56 h of run time per battery recharge, it offers a good week of operation on its own power, or will operate continuously with its charger plugged in. When closed, the unit is relatively watertight. Temperature tests, which were run on the most thermally sensitive parts of the system, indicate a conservative operating range of 30 to 100^oF.

While this unit was primarily intended to measure ¹³⁷Cs, it can, with not much effort, be modified to measure a variety of other monoenergetic gamma rays.