Characterization of the neutron detector upgrade to the GP-SANS and Bio-SANS instruments at HFIR

Abstract

Over the past year, new 1 m×1 m neutron detectors have been installed at both the General Purpose SANS (GP-SANS) and the Bio-SANS instruments at HFIR, each intended as an upgrade to provide improved high rate capability. This paper presents the results of characterization studies performed in the detector test laboratory, including position resolution, linearity and background, as well as a preliminary look at high count rate performance.

Introduction

The High Flux Isotope Reactor (HFIR) is an 85 MW research reactor located at the Oak Ridge National Laboratory in Oak Ridge, Tennessee. Though its original mission was focused on transuranic isotope production, HFIR's potential for neutron scattering was recognized at the outset [1], and neutron beam tubes were installed during its initial construction in the 1960s. HFIR is still actively involved in isotope production, as well as materials irradiation and neutron activation studies, while its role as a center for neutron scattering science has increased in prominence over the years.

The reactor is now in its fifth decade of operation as a bright source of thermal neutrons for scattering experiments. A liquid hydrogen cold source was installed and commissioned in 2007, extending neutron wavelengths into the 4–12 Å range. With a peak thermal neutron flux greater than 2×10¹⁵/cm²/s, HFIR has the highest flux of any reactor-based neutron source in the United States, and a cold source flux comparable to the best in the world.

The Cold Guide Hall was built to house a suite of instruments specifically designed to take advantage of this new high flux, cold neutron beam. The first two of these beamlines to be commissioned are dedicated to Small Angle Neutron Scattering (SANS): the general purpose GP-SANS (CG-2) and Bio-SANS (CG-3). SANS instruments provide information on the structure and dynamics of materials with relatively large d spacing, ranging from a few up to several thousand Angstroms. GP and Bio-SANS instrument operating parameters can be found on the ORNL website [2].

In a typical reactor-based SANS instrument [3], source and sample apertures are combined with a velocity selector to produce a well collimated, monochromatic beam of neutrons at the sample position. A 2-dimensional position sensitive detector is located at some distance (often meters) downstream from the sample, its active area normal to the beam direction, to record the forward scattered intensity distribution. The sample-to-detector distance, as well as the detector itself, is located inside a vacuum chamber to minimize air scatter in the secondary flight path. The 2d or area detector rides on a track, allowing translation along the beam direction, thus enabling measurements to be made over a wide range of scattering angles.

The forward scattered flux very near the sample can be quite high; conversely, a SANS experiment may require the use of relatively weak scattering samples. This produces the seemingly contradictory detector requirements of the need to operate at both very high and very low count rates. The low rate condition dictates an important criterion in the choice of position sensitive detector to be selected for use in a SANS instrument – it must have low background, and, in particular, low gamma-ray sensitivity. Gas proportional detectors using ³He as the neutron converter are extremely insensitive to gammas (10⁻⁷ typical), owing in part to the low Z of helium, and for this reason are used almost exclusively in SANS instruments worldwide [4]. Operation of gas proportional detectors at high input rates is limited by the saturation effects associated with space charge accumulation [5], and long term rate related performance degradation can occur in some cases due to the formation of deposits on the anode wires [6].

Both of the SANS beamlines at HFIR were originally instrumented with identical commercial ³He filled multiwire proportional chambers of 1 m×1 m active area. These detectors consist of an anode wire plane at positive high voltage, sandwiched in between two orthogonally oriented cathode wire planes, which are used for signal readout. The readout utilizes the amplifier per wire approach, relying on XY coincidence for position determination. Cathode wire pitch and hence position resolution is approximately 5 mm in both the X and Y directions. Although the detection efficiency, position resolution, and gamma sensitivity were sufficient for this instrument, it was realized early on that the maximum practical count rate of these detectors was limited to approximately 20,000 cps over the entire array, or around 100 cps per wire. This was far lower than expected, and therefore a significant limitation to the instrument capability.

The GP and Bio-SANS detectors have recently been replaced with a completely different type of design, intended as an upgrade with improved count rate performance. This detector is modeled after the system currently installed in the EQ-SANS instrument at ORNL's Spallation Neutron Source [7], and is also quite similar to the 1 m×1 m detector which has been operating with MHz rate capability at the SANS beamline D22 of the Institut Laue-Langevin (ILL) for almost 10 years [8].

Indeed, tests indicate that these new detectors are performing as expected, collecting global rates at least two orders of magnitude greater than those of the original system, with little or no indication of saturation effects, and no evidence of permanent damage.

This paper presents some of the key performance characteristics of this detector upgrade. In Section 2, the physical properties of the detector are described, followed by a discussion of basic operating principles in Section 3. Section 4 details the various characterizations tests performed in the detector laboratory, at relatively low count rate. Finally, high rate test results are covered in Section 5.