Compton Suppression Systems

Figure 6. An ORTEC Compton Suppression System.

The large Compton background seen in Fig. 5 is due to capturing only part of the original gamma-ray energy. Some of the remaining or lost energy actually leaves the detector. If we could capture these lost energy parts, we could reject the part of the gamma-ray energy captured in the Germanium. In a Compton Suppression System (CSS), (Fig. 6) the Germanium detector is surrounded by a NaI(Tl) annular detector which detects photons escaping from the Germanium. The signals from the Germanium and NaI are used in anticoincidence circuitry to remove the Compton background events from the primary gamma-ray spectrum. CSS systems are usually tested with 137Cs, and some systems have background improvements of a factor of 5 over the p/C ratio of the detector itself.

CSS Disadvantages: The sample size is usually small because the sample must be placed inside the NaI annulus. The system is complex: coincidence electronics require careful adjustment and maintenance to ensure consistent performance. CSS efficiency calibration is complex because some nuclides naturally emit photons in coincidence with each other and this reduces the full-energy peak areas for these peaks as well as reducing the background. Most importantly CSS systems are expensive. It is usually better to spend your money on a larger Germanium (HPGe) detector, which may be less expensive, work better, and be simpler than Compton suppressing a smaller detector. Compton suppression of large detectors is not as effective (not as large an improvement) as small detectors because the large detectors already have a high p/C ratio. High p/C Germanium (HPGe) detectors have largely replaced CSS.

Detector Efficiency: ε(E)

Figure 7. Point source efficiency curves for planar and coaxial detectors in arbitrary units.

Figure 8. Three different crystal geometries, all with the same IEEE point source efficiency, but very different absolute efficiency for a puck sample on endcap.

The IEEE-325 definition of relative efficiency (Ref. 4) at 1.33 MeV, is not a good indicator of detector sensitivity in most of the sample geometries you want to use. It is defined at a single energy and for a point source at 2 cm distance to the detector endcap. No real samples meet this criteria except a point ⁶⁰Co source at 25 cm from the endcap! On the other hand, relative efficiency is often a good place to start as a general indicator of detector performance. The efficiency for various energies is shown in Figure 7.

Figure 8 illustrates how three detectors can have the same IEEE-325 efficiency, yet have different efficiencies for your samples and your nuclides. All three schematically represented detectors have the same IEEE-325 relative efficiency, but for counting a flat disk-like sample (e.g., filter paper), it is obvious that the long and thin detector will have poorer geometrical efficiency than the "shorter and fatter" detectors. So if your samples are filter papers, disks or other large area containers, your best selection will be a shorter and fatter detector, such as the ORTEC PROFILE GEM. With the ORTEC PROFILE GEM series, you can specify the crystal dimensions as well as IEEE-325 relative efficiency.