Model-based design evaluation of a compact, high-efficiency neutron scatter camera

Abstract

This paper presents the model-based design and evaluation of an instrument that estimates incident neutron direction using the kinematics of neutron scattering by hydrogen-1 nuclei in an organic scintillator. The instrument design uses a single, nearly contiguous volume of organic scintillator that is internally subdivided only as necessary to create optically isolated pillars, i.e., long, narrow parallelepipeds of organic scintillator. Scintillation light emitted in a given pillar is confined to that pillar by a combination of total internal reflection and a specular reflector applied to the four sides of the pillar transverse to its long axis. The scintillation light is collected at each end of the pillar using a photodetector, e.g., a microchannel plate photomultiplier (MCP-PM) or a silicon photomultiplier (SiPM). In this optically segmented design, the (x, y) position of scintillation light emission (where the x and y coordinates are transverse to the long axis of the pillars) is estimated as the pillar's (x, y) position in the scintillator "block", and the z-position (the position along the pillar's long axis) is estimated from the amplitude and relative timing of the signals produced by the photodetectors at each end of the pillar. The neutron's incident direction and energy is estimated from the (x, y, z)-positions of two sequential neutron-proton scattering interactions in the scintillator block using elastic scatter kinematics. For proton recoils greater than 1 MeV, we show that the (x, y, z)-position of neutron-proton scattering can be estimated with < 1 cm root-mean-squared [RMS] error and the proton recoil energy can be estimated with < 50 keV RMS error by fitting the photodetectors' response time history to models of optical photon transport within the scintillator pillars. Finally, we evaluate several alternative designs of this proposed single-volume scatter camera made of pillars of plastic scintillator (SVSC-PiPS), studying the effect of pillar dimensions, scintillator material (EJ-204, EJ-232Q and stilbene), and photodetector (MCP-PM vs. SiPM) response vs. time. We demonstrate that the most precise estimates of incident neutron direction and energy can be obtained using a combination of scintillator material with high luminosity and a photodetector with a narrow impulse response. Specifically, we conclude that an SVSC-PiPS constructed using EJ-204 (a high luminosity plastic scintillator) and an MCP-PM will produce the most precise estimates of incident neutron direction and energy.

Interaction position, time, and energy resolution in organic scintillator bars with dual-ended readout

“…The methods used here are based on the difference in the pulse arrival time and amplitude at each end of the bar. Potentially, the difference in pulse shape provides an additional method [1], however this is not explored in this work. In the case of the pulse amplitude difference, a shorter attenuation length maximizes the difference in amplitude and therefore the position resolution.…”

“…In this instance of the concept, only the position along each bar segment of the array must be reconstructed for the position of each interaction, while the other position dimensions are determined by the particular photodetector in which the scintillation light was detected. Monte Carlo simulations of various detector configurations of optically segmented bars have been previously performed [1] and provided guidance as to the optimal choice of scintillator, reflector material, and photon detection technology. However, practical considerations have limited the experimental space explored.…”

Proton light yield in organic scintillators using a double time-of-flight technique

“…For example, neutron imaging techniques using double-scatter kinematic reconstruction depend on the proton light yield relation to determine the energy of the recoil proton following an n-p elastic scattering event and thereby infer the angle of the incident neutron. [1,2] The light yield relation is also used as input to Monte Carlo simulations of the response of neutron detection systems, which in turn is a critical input for neutron spectroscopy techniques using pulse height spectrum unfolding. [3,4] Several categories of methods exist for the measurement of proton light yield-direct methods, indirect methods, and edge characterization techniques-as detailed in Sec.…”

The Architected Multi-material Scintillator System: Designs and Modeling

“…These cuts applied left the simulated detector array with a 12.6% neutron detection efficiency among all neutrons that pass through the scintillator. This 12.6% detection efficiency is significantly higher than projected for neutron scatter cameras, which have neutron detection efficiency between 0.1-1% [7] due to requiring two above-threshold scatters for each neutron. With these two cuts, 10 6 neutrons were used to train the analysis model.…”