Digital Pulse Shape Analysis with Phoswich Detectors to Simplify Coincidence Measurements of Radioactive Xenon

The Comprehensive Nuclear-Test-Ban Treaty establishes a network of monitoring stations to detect radioactive xenon in the atmosphere from nuclear weapons testing. One such monitoring system is the Automated Radioxenon Sampler/Analyzer (ARSA) developed at Pacific Northwest National Laboratory, which uses a complex arrangement of separate beta and gamma detectors to detect beta-gamma coincidences from the xenon isotopes of interest. The coincidence measurement is very sensitive, but the large number of detectors and photomultiplier tubes requires careful calibration. Simplifying this coincidence measurement system while maintaining its performance is the objective of the research described here. It has been suggested that beta-gamma coincidences could be detected with only a single photomultiplier tube and electronics channel by using a phoswich detector consisting of optically coupled beta and gamma detectors (Ely, 2003). In that work, rise time analysis of signals from a phoswich detector was explored as a method to determine if interactions occurred in either the beta or the gamma detector or in both simultaneously. However, this approach was not able to detect coincidences with the required sensitivity or to measure the beta and gamma energies with sufficient precision for radioxenon monitoring. In this paper, we present a new algorithm to detect coincidences by pulse shape analysis of the signals from a BC404/CsI(Tl) phoswich detector. Implemented on fast digital readout electronics, the algorithm achieves clear separation of beta only, gamma only and coincidence events, accurate measurement of both beta and gamma energies, and has an error rate for detecting coincidences of less than 0.1%. Monte Carlo simulations of radiation transport and light collection were performed to optimize design parameters for a replacement detector module for the ARSA system, obtaining an estimated coincidence detection efficiency of 82-92% and a background rejection rate better than 99%. The new phoswich/pulse shape analysis method is thus suitable to simplify the existing ARSA detector system to the level of a single detector per sample chamber while maintaining the required sensitivity and precision to detect radioxenon in the atmosphere. Ely, J. H. et al (2003), “NOVEL BETA-GAMMA COINCIDENCE MEASUREMENTS USING PHOSWICH DETECTORS” in Proceedings of the 25th Seismic Research Review – Nuclear Explosion Monitoring: Building the Knowledge Base, LA-UR-03-6029. OBJECTIVE The Comprehensive Nuclear-Test-Ban Treaty establishes a network of monitoring stations to detect radioactive xenon in the atmosphere from nuclear weapons testing. One such monitoring system is the Automated Radioxenon Sampler/Analyzer (ARSA) developed at Pacific Northwest National Laboratory (Reeder, 1998). The ARSA system consists of a pair of large NaI(Tl) scintillator crystals holding four cylindrical fast plastic scintillator (BC-404) cells which are optically isolated from the NaI(Tl). The cells are filled with the xenon gas to be counted, which decays by emitting gamma rays or X-rays in coincidence with beta particles or conversion electrons. The plastic scintillator is meant to absorb all beta particles and conversion electrons, while the longer range gamma rays and X-rays will mainly be absorbed in the NaI(Tl) scintillator. Each BC-404 cell and each NaI(Tl) crystal is coupled to a pair of photomultiplier tubes (PMTs) and is read out by independent electronic channels. The sensitivity for detecting xenon isotopes is greatly increased by requiring coincidence between the signals from the PMTs coupled to the NaI(Tl) and the signals from the PMTs coupled to the BC-404. While obtaining high coincidence detection efficiency and resolutions of about 25% for characteristic 80keV gamma rays, the current ARSA system design results in significant operational complexity. The principle of time based coincidence, while effective in suppressing the background, requires separate signals from the NaI(Tl) and the BC404, i.e. separate PMTs and readout electronics. In particular, the 12 PMTs require careful gain matching and calibration, and as the PMT gains change with time, voltage and temperature, the system easily drifts out of calibration. To improve the current ARSA system a new concept for coincidence measurements was explored previously (Ely, 2003) based on rise time analysis of signals from a phoswich detector. The phoswich detector consisted of a 0.04 inch thick CaF2(Eu) crystal (decay constant 940ns) optically coupled to a 2x2 inch NaI(Tl) crystal (decay constant 250ns) and was read out by a single PMT. The CaF2(Eu) is used as the beta detector to absorb beta particles and conversion electrons and the NaI(Tl) acts as the gamma ray detector to absorb gamma rays and X-rays. By integrating the signal from the PMT in a charge integrating preamplifier and acquiring pulse waveforms with a fast digital pulse processor, the scintillator in which radiation interacted could be determined by the signal rise times. While this method of pulse shape coincidence detection worked well to distinguish CaF2(Eu) only events and NaI(Tl) only events, coincident events in both scintillators were not easily identified by this algorithm and/or choice of scintillators and it was deemed challenging to separate the individual gamma and beta contributions with any precision. In this paper, we describe an improved method to detect coincidences with a phoswich detector and single channel of readout electronics. The method uses the signal directly from the phoswich detector, without a charge integrating preamplifier, and determines the scintillator(s) in which the interaction occurred by analyzing the signal over characteristic time periods, thus detecting coincidences. Applied to radioxenon monitoring, the method provides beta-gamma coincidence detection and energy measurement of both beta and gamma energies with a greatly simplified measurement setup. RESEARCH ACCOMPLISHED 1. Development of Pulse Shape Analysis with Prototype Phoswich Detector The pulse shape analysis algorithms were developed using a prototype phoswich detector consisting of a 1” diameter by 1” thick CsI(Tl) crystal optically coupled to a 1” diameter, 1mm thick disk of the plastic scintillator BC-404 on the front end. The detector was illuminated with a variety of solid sources or Xe and Rn gas enclosed in small plastic bags. Using an XIA Pixie-4 digital spectrometer directly connected to the PMT coupled to the detector, we acquired waveforms of the detector signals and found the three basic types of events shown in Figure 1: a) slow rising and slow falling pulses corresponding to interactions only in the CsI, b) very fast pulses with high amplitude corresponding to interactions only in the BC-404, and c) combinations of the previous cases corresponding to coincident interactions in both scintillators. Limiting the signal bandwidth of the Pixie-4 analog front end reduced the amplitude of the fast BC-404 signals without affecting the slower, low amplitude CsI signals. A single channel could thus accommodate the highest beta particle energies expected from the radioxenon decays while still obtaining sufficient precision for low amplitude X-ray signals.