Low Background Germanium Detectors - Material Selection and Measurements

Low-Background Germanium (HPGe) Detectors

Material Selection and Low-Background Measurements

For many years ORTEC has been selecting low-background materials such as magnesium, ultra-pure aluminum, OFHC copper, lead, stainless steel, beryllium (for windows), and charcoal (for cryopumping).

ORTEC has also established a low-background spectroscopy laboratory and a detector background-certification program. The laboratory houses two lead shields, configured to accommodate detectors in various configurations. Each shield consists of a 3-in. OFHC copper "well" surrounded by eight inches of low-radiogenic lead. The copper virtually eliminates the lead x-rays resulting from photoelectric interactions of gamma rays with the lead.

For characterizing the materials to be used for detector cryostat construction, the low-background laboratory has a graded-Z (Pb, Cu, Cd) shield containing an ultra-low-background detector. Quality control limits have been established for each type of material dependent on its location in the finished cryostat.

Each completed detector is placed in the appropriate shield. An NBS SRM 4275 source is used for calibration.

Then, with the source removed, a detector background spectrum is acquired for 100,000 seconds. The background spectrum is searched using a second-difference peak-search algorithm. All identified peaks are visually examined. For any peaks that are part of the list reported as "not found," a region-of-interest (ROI) is set manually, and the net area is computed by the MCA Emulation software. Finally, a report that lists the ROI net count rates is created.

The data are reported as intensities rather than activities because the activities are a function of the geometry of the calibration, while the intensities are geometry-independent.

The logarithmic plot (Fig. 21), linear plot (Fig. 22), and low-background analysis reports (Tables 9 and 10) show, for comparison, the gamma background spectra of two GEM detectors of identical relative efficiency (56%), one in a standard cryostat and the other in an extra-low-background (Model XLB) cryostat (Fig. 23). The difference in background is substantial. The background report can be supplied upon request, and can be used in conjunction with the ORTEC MAESTRO MCA Emulation program to visually examine the data or to plot it.

It must be emphasized that the gamma background measured in a detector cannot be better than the background in the laboratory where the detector is operated. For example, health physicists making in vivo measurements should be aware that the beds or chairs in which the subject is placed are generally not made of materials selected for low gamma background and, also, that the subjects being measured emit gamma rays at a rate consistent with the 40–120 nanocuries of ⁴⁰K normally found in the human body. Therefore, while every precaution must be taken to obtain good, reliable measurements, it is useless to strive for the "ultimate in low background" as some physics researchers operating in sophisticated underground laboratories must and can do.

The data obtained at ORTEC are representative of the low-background characteristics of the detector/shield combination in the ORTEC facility in Oak Ridge, Tennessee. Since background results are dependent upon the shielding, better results may be obtained in sophisticated laboratories (e.g., with active shields or in deep mines).

An example of this for a 40% GEM in a vertical XLB cryostat is shown in Table 11. The third column is from the report of the ORTEC Low-Background Facility. The fourth column contains the reported results on the same detector in a sophisticated low-background laboratory (Ref. 13). Another example is given in Table 12, which reports results obtained at the Gran Sasso National Laboratory, an Italian research facility located under 3500 meters of rock, with a resulting cosmic background reduction of a factor of 10⁶. The ORTEC detector has a measured efficiency of 96%. Extraordinary precautions were taken to minimize the gamma background.^*

When comparing the data in Tables 9, 10, and 11, the following considerations must be kept in mind:

As there is no standard, laboratories use different formats for such tables; hence, the obvious differences between Tables 9 and 10 (both obtained at ORTEC) and 11 (obtained at the U.S. National Institute of Standards and Technology).

The computer printout reports energy centroids rather than the energy of the nuclides. Therefore, some interpretation is required to understand the centroid/nuclide relationship between Tables 9 and 10 and Table 8. For example, Region #1 in Table 9 (centroid at 45.72 keV) indicates the counts due to ²¹⁰Pb in Table 8 (46.5 keV).

When comparing gamma background data obtained from detectors with different efficiency, the difference in efficiency should be factored in, at least in an approximate (linear) way.

A less sophisticated way of characterizing low-background detectors is reporting the total counts per second in a given energy interval, typically from 100 keV to 3 MeV. A large (96% efficiency) ORTEC detector measured at Gran Sasso, the world-class Italian laboratory under a mountain, registered 100 counts per second in that energy interval.

*C.R. Arpesella, et al., "A Low Background Counting Facility at Laboratori Nazionali del Gran Sasso." (Internal Report LNGS – 92/35 July 1992).