Efficiency of Si(Li) and Ge X-ray Detectors
Depending upon the approach taken to standardization in PIXE, the efficiency of the detector may be needed in either absolute or relative terms. Over many years we have published a number of papers on this topic, and our approach to efficiency modeling is very similar to those of other authors such as D.D. Cohen or W. Maenhaut.
Recently, however, we have made changes to our efficiency model. It has been customary to neglect contributions to the full energy peak that arise from one or more Compton scattering events followed by a photo-electric absorption. With Dr. Joanne O'Meara, we have used Monte Carlo modeling to show that this is not a valid approximation above about 25 keV, and we have presented a simple means of including this effect in efficiency calculations. At the same time, we have made our entire calculation more rigorous. To test our new approach we deduced the thickness of a particular Gresham Scientific Instruments detector from its measured and fitted efficiency curve. The value obtained corresponded closely with the value from the manufacturing records, suggesting that our modeling was indeed reliable.
In a different paper we considered the widely-used scheme of scanning collimated photon beams across a detector and along its axis to determine its position and dimensions. Pitfalls in this approach were demonstrated and means to ensure accuracy were presented and demonstrated.
We agree with the arguments made in the papers of Dr. Tibor Papp that insufficient attention has been paid to the effect of the signal processing electronics on the efficiency. More work needs to be done on this topic using different analog and digital processors.