Comparing the response of PSD-capable plastic scintillator to standard liquid scintillator

Abstract

Listen

Translate article

This work discusses a test campaign to characterize the response of the recently developed plastic scintillator with pulse shape discrimination (PSD) capabilities (EJ-299-33). PSD is a property exhibited by certain types of scintillating material in which incident stimuli (fast neutrons or γ rays) can be separated by exploiting differences in the scintillation light pulse tail. Detector geometries used were: a 10cm×10cm×10cm cube and a 10-cm diameter×10-cm long cylinder. EJ-301 and EJ-309 liquid scintillators with well-known responses were also tested. The work was conducted at the University of Massachusetts Lowell Van De Graaff accelerator. The facility accelerated protons on a thin Li target to yield quasi-monoenergetic neutrons from the 7Li(p,n)7Be reaction (Q-value: –1.644MeV). Collimated fast neutrons were obtained by placing detectors behind a neutron spectrometer. Rotating the spectrometer, and thus changing the neutron energy, allowed us to achieve 0.5–3.2MeV neutrons in 200–300keV steps. Data were acquired through a flash analog-to-digital converter (ADC) capable of performing digital PSD measurements. By using the PSD technique to separate the neutron events from unwanted γ background, we constructed a pulse height spectrum at each energy.Obtaining a relationship of the relative light output versus energy allowed us to construct the response function for the EJ-299-33 and liquid scintillator. The EJ-299-33 response in terms of electron equivalent energy (Ee.e.) vs. proton equivalent energy (Ep.e.), how it compared with the standard xylene-based EJ-301 (or, NE-213/BC-501A equivalent) and EJ-309 liquid scintillator response, and how the EJ-301 and EJ-309 compared, are presented. We find that the EJ-299-33 demonstrated a lower light output by up to 40% for <1.0MeV neutrons; and ranging between a 5–35% reduction for 2.5–3.0MeV neutrons compared to the EJ-301/309, depending on the scintillator and geometry. Monte Carlo modeling techniques were used to investigate how the neutron beam and accelerator background environment affected the detector response. We find relatively good agreement between our results and the modeling; however, the observed response could not be fully accounted for due to events with pulse pile up, thus leading to contamination of the neutron PSD selected events.