Efficient Reduction of Piled-Up Events in Gamma-Ray Spectrometry at High Count Rates

When dealing with high count rate γ-ray spectrometry, the pile-up phenomenon is an important obstacle to high detection efficiency and good energy resolution. In traditional pulse processing piled-up events are rejected to preserve good energy resolution at the expense of detection efficiency. This is especially problematic in in-beam γ-ray spectrometry, where searching for coincidences between two different detection signals further accentuates the loss of efficiency. Consequently we want to detect as many events as possible and obtain the best possible energy resolution. We propose a method that reduces the number of piled-up events, using a parallel moving window deconvolution (MWD) block implementation where the shaping time parameter differs for every MWD block. With appropriate analysis logic we get more experimental information through reduced dead time, at the cost of controlled and selectively worsened energy resolution, on an event-by-event basis, thus achieving better overall detection efficiency. We tested our method on real experimental data for different count rates. At 500 kcps count rate the detection efficiency of our method is higher, by a factor of about 37, than the efficiency of a standard digital method commonly used in commercial systems with pile-up rejection.