Characterisation of a position sensitive thermal neutron detector

Boyd, Laura (2021) Characterisation of a position sensitive thermal neutron detector. PhD thesis, University of Glasgow.

Full text available as:
Download (14MB) | Preview


New types of position sensitive thermal neutron detectors for use in neutron imaging are in high demand due to the rising cost of 3He used in most existing thermal neutron counters. A novel 6Li doped scintillation glass sheet (GS20) coupled to a multi-anode photomultiplier tube with a portable data acquisition system, known as the Solid State Neutron Detector (SoNDe), is one possible solution. This study focuses on the characterisation and understanding of the Solid State Neutron Detector for application at the European Spallation Source (ESS), where the Small Angle Neutron Scattering instrument requires approximately 400 of these modular detectors. Therefore, understanding the efficiency, crosstalk, gain non-uniformities, and response to radiation of the SoNDe module is crucial.

Radiation from a number of sources were used to establish the main characteristics of the detector and the results compared. Sources included a 404 nm laser, 241Am producing alpha particles, AmBe producing fast and moderated neutrons as well as gamma-rays, 60Co and 137Cs producing gamma-rays, and a pelletron accelerator producing protons. A simulation written using the Geant4 toolkit, verified through comparison to measured data, was used to study the details of the detector construction and to predict the response of the detector to possible sources of background radiation at ESS. The gains and gain stability of the photomultiplier pixels are important for operation at ESS.

The relative pixel gains of the photomultiplier tube were provided by Hamamatsu for each individual photomultiplier. However, given the importance of these numbers to determining detector response and setting readout thresholds, they were also measured using various methods. These methods included the use of a collimated alpha source and a laser spot centred on each pixel separately, a uniform square diffuse laser beam simultaneously illuminating all pixels, and a moderated neutron source illuminating all pixels. The alpha, neutron, and diffuse laser results are similar to the Hamamatsu measurement and give potential methods of detector calibration. The laser spot data shows a slightly different gain, and is less directly comparable to the production measurement.

The gain stability of the detector system and its drift characteristics were measured using the collimated laser and alpha-particle sources. The 404 nm laser was close in wavelength to the peak in the scintillation emission spectrum of the GS20 glass. Drifts of 10−4 %/hr were found. This indicated a minute shift in gain in the detector over an extended period of time (34 - 67 hours). The stability studies also highlighted the importance of a sufficient warm-up period, 1 hour in duration, prior to operation.

Various sources were scanned across the detector face to provide an in-depth look at the position dependence of the pulse-height response of the detector. Fine scans over the entire detector area were performed with a collimated laser beam at single and 60 photoelectron light intensity levels. The results revealed areas of higher/lower response (dead spaces) some of which is due to the presence of photomultiplier tube structures, such as focusing electrodes between the cathode and first dynode.

The crosstalk between pixels was measured using both the collimated alpha source and the laser. The percentage of signal in the neighbouring pixels surrounding the target pixel was measured. A range of 0.0 - 3.0 % was measured for the laser data, and 2.5 - 9.8 % for the alpha data. Both data sets were greater than the Hamamatsu results of 0.0 - 2.1 %. While the laser light was highly directional the scintillation light was emitted isotropically and gave rise to the increased crosstalk percentages in the surrounding pixels. The spreading of scintillation light from its point of origin was investigated in more detail with the Geant4 simulation.

Cross-pixel scans in the horizontal, vertical, and diagonal directions were carried out with collimated alpha particle and proton beam sources. The alpha beam spot on target was ∼ 1 mm diameter, while the proton beam spot was ∼ 0.1 mm diameter permitting finer scanning steps. The measured results revealed the variation in signal amplitude as the source was scanned across the detector face with the proton data presenting an especially stringent test for the predictions of the Geant4 based model of the detector.

The Solid State Neutron Detector’s ability to successfully detect thermal neutrons is highly important. A moderated AmBe source was used to irradiate the detector. A clear neutron peak was distinguished. A Geant4 simulation was used to compare with the measured data, and simulate possible neutron/gamma backgrounds with energies ranging from 0.025 eV - 100.0 MeV.

The Solid State Neutron Detector has been thoroughly characterised in stability, relative gain, and response to various forms of radiation. The detector optimisation revealed a number of possible alternative se tups, with the originally proposed setup proving the most effective.

Item Type: Thesis (PhD)
Qualification Level: Doctoral
Subjects: Q Science > QB Astronomy
Q Science > QC Physics
Colleges/Schools: College of Science and Engineering > School of Physics and Astronomy
Supervisor's Name: Seitz, Dr. Bjoern and Montgomery, Dr. Rachel and Annand, Dr. John
Date of Award: 2021
Depositing User: Theses Team
Unique ID: glathesis:2021-82291
Copyright: Copyright of this thesis is held by the author.
Date Deposited: 30 Jun 2021 13:55
Last Modified: 30 Jun 2021 13:55
Thesis DOI: 10.5525/gla.thesis.82291

Actions (login required)

View Item View Item