Neutron Interaction Properties of Structural Materials for Multi-Grid Neutron Detectors

TL;DR

This work addresses the mitigation of internal neutron scattering in the T-REX Multi-Grid detector by evaluating shielding strategies built from $\mathrm{B_{4}C}$-containing materials. The authors combine transmission measurements on the EMMA beam line with scattering measurements on Merlin across a range of shielding candidates, comparing results to Geant4 simulations and a simple model $R_{io}=\exp(-k\lambda_n)$ with $k=2133 N_{10} t$. They find that uncoated $\mathrm{B_{4}C}$/Al composites provide the strongest attenuation and lowest scattering, while Ni plating reduces alpha background but can increase scattering; thicker or less uniform coatings generally perform worse. A TRP-3 prototype was built using NiBA31-1 radial blades and BA31 shielding, and initial tests indicate substantially reduced internal scattering, informing shielding choices for accurate energy and momentum transfer measurements in sub-eV neutron spectroscopy at ESS/ISIS.

Abstract

The T-REX neutron time-of-flight spectrometer at the European Spallation Source will use Multi-Grid Technology, which relies on thin B4C coatings on the Al blades of the grids to detect scattered thermal neutrons. Following a Monte Carlo study of internal shielding to suppress neutron multiple scattering in T-REX, the neutron transmission and scattering properties of 12 shielding-material samples have been measured at the ISIS spallation neutron source. Neutron transmission was measured on the EMMA beam line at wavelengths 0.5-4.7 A, using a 2D-position-sensitive, neutron GEM detector, while neutron scattering was measured for 6 of the samples at the Merlin spectrometer, at wavelengths 0.72, 1.28, 1.85 and 2.41 A. The present tests show that a B4C/Al composite material, plated with Ni to stop intrinsic alpha background, is an effective neutron absorber, suitable for incorporation in the Multi-Grid structures which detect the neutrons in inelastic neutron spectrometers .

Figures (10)

• Figure 1: Left: prototype T-REX grid. Each voxel has an internal dimension of 23.5 mm (x) by 24.0 mm (y) by 9.5 mm (z). Right: 3D view of a 12-grid T-REX prototype, with outer sections cut away to reveal the internal structure.
• Figure 2: Left: schematic diagram of the EMMA neutron transmission test. Right: comparison of sample in/out wavelength spectra derived from TOF at the nGEM detector. The sample was BA31-3 (see Table \ref{['tab:Shielding-candidates']}).
• Figure 3: Average ratio $R_{io}$. The data are displayed in black and the red lines are exponential fits to the $R_{io}$ data. Geant4-simulated distributions (G4) are shown in blue.
• Figure 4: 2D images of neutron absorption $1-R_{io}$ on eight samples. Note that the vertical scales of the plots are different to highlight any variations across the samples. Horizontal black dotted lines (plot BA31-3) show the limits for the X-projections in Fig. \ref{['fig:1D-slices-of']}, while the vertical red dotted lines show the Y-projection limits.
• Figure 5: 1D projections of the 2D neutron absorption plots in Fig. \ref{['fig:2D-images']}, with X-projections (black) summed over y bins 60-64 and Y-projections (red) summed over x bins 66-70. Note that the Y-axis scales are different as $1-R_{io}$ varies considerably from sample to sample.