Effect of an Occulter on a Cobalt-57 Spectrum

Rev. C (25 March 2002):

Included rebinned plots of the spectral data. (Rev. B)
Revised conclusion as to affect of lead shielding on fitting. (Reb. B)
Revised Section 4 to reflect correction of attenuation calculation.

1. Introduction

The purpose of this experiment was to find the difference in the measured spectrum between two test setups: one with an occulter and one without. In both setups, a Cobalt-57 source was placed inside a lead box with a single CZT detector (see the drawing in http://lheawww.gsfc.nasa.gov/~derek/mtfit/).

2. Experimental Procedures

As in the previous experiment ( http://lheawww.gsfc.nasa.gov/~derek/mtfit/), I placed a Cobalt-57 source approximately 13 cm from the detector holder. The entire setup was enclosed inside 2 mm thick lead shielding with a small circular opening for the end of the brass detector holder and a small square hole behind the source The purpose of the small square hole was to reduce backscattering--photons that would have backscattered (if the hole had not been present) would instead exit the setup, and the probability that they would scatter off of outside materials and re-emerge through the hole would be small.

The occulter consisted of a small lead square, cut from a BAT mask tile, 1mm thick and between 1.5 and 1.8 mm on a side. It was taped directly over the active area of the source (onto the plastic disk housing the source. Figures 1, 2, and 3 show scale drawings of the test setup:

Figure 1:
Test Setup
(Top View)

Figure 2:
Test Setup
(Side View)

Figure 3:
Test Setup
(Close-Up of Occulter and Source)

As shown in the diagram, the opening angle of the occulter was between 16° and 24°. That means the occulted region at the detector was between 3.4 and 5.3 cm wide.

Detector I44-609317-F4 was used. Each spectra was taken for 30 minutes (live time). A lead shield was placed about 12 inches behind the source to protect the rest of the lab and those working in it. The bias voltage was 300 V.

3. Results

Shown below are the full channel range spectra without the occulter, with the occulter, and the difference between the two:

Figure 4:
Full Channel Spectra

Figure 5:
Full Channel Spectra (Rebinned)

The subtracted spectrum is the expected spectrum from a cobalt-57 source. The bump at around channel 2250 is consistent with the expected escape peak features of the spectrum.

Shown below are the spectra for the channel ranges used in the fitting for mu-tau holes described in http://lheawww.gsfc.nasa.gov/~derek/mtfit/:

Figure 6:
Upper Channel Spectra

Figure 7:
Upper Channel Spectra (Rebinned)

One thing one notices right away is that the occulter increases the height of the tail by about 5% or so. This might be enough to affect the fit value of (μτ)_h when different channel regions are used. This is something that should be checked.

4. An Aside: Expected Transmission

The peak heights of the 122 keV and 136 keV peaks for the occulter and no-occulter spectra were:

This implies transmissions of:

The measured thickness of the occulter is 0.94 ± 0.03 mm, whereas this transmission is consistent with a thickness of 0.81 mm. One possible explanation for this discrepancy comes to mind. The occulter was not square at the edges, but rather tapered. It the photons passed partially through one of those edges, the overall effective thickness would be lower.

5. Conclusions

The affect of the lead shield on the portion of the spectrum used to determine (μτ)_h might be enough to change affect the fitting. The transmission through the occulter is less than expected--consistent with a thickness of about twice the actual thickness.

Return to Home Page