Search

The most compelling explanation for the so-called Dark Matter of the Universe is the postulation of particles beyond the standard model, with Weakly Interacting Massive Particle (WIMP) dark matter being well-motivated. While there are many different methods to search for WIMPs, the most sensitive dark matter experiments in the world employ liquid noble gas targets to detect WIMP-induced recoils. As the next generation of liquid noble detectors become more sensitive, they are confronted by an inevitable background of solar neutrinos, which inhibit the conclusive identification of dark matter in such searches. Directional dark matter detectors have the capability to distinguish against the otherwise irreducible solar neutrino background by adding information about the direction of the WIMP-induced recoil events. Most directional detectors reconstruct recoil tracks using low-pressure gas Time Projection Chambers (TPC). In gas TPC operation, it is important to remove radon and common pollutants from the target gas. Radon contamination provides a source of unwanted background able to mimic WIMP-induced recoils, while common pollutants can significantly suppress the gain of the detector. SF6 is an ideal target gas for directional dark matter searches, so the ability to remove radon and common pollutants from SF6 during TPC operation is crucial. A method that also recycles SF6 is required as it is a potent greenhouse gas. This thesis describes work toward a gas recycling system that removes radon and common pollutants from target gases during TPC operation. The removal of radon from SF6 gas was demonstrated for the first time using a 5 angstrom type molecular sieve. A low radioactive 5 angstrom type molecular sieve that intrinsically emanated 98.9% less radon per radon captured compared to commercial sieves was found. To effectively implement the molecular sieves with TPC detectors, a gas system utilising a modified Vacuum Swing Adsorption (VSA) technique with a gas recovery buffer was designed. The VSA technique minimises the required amount of molecular sieve for long-term filtration, and the gas recovery buffer maximises the amount of recycled gas. The design was built into a prototype and tested with a small-scale gas TPC detector. Performance testing with the gas system prototype resulted in the low radioactive 5 angstrom type molecular sieve reducing the intrinsic radon contamination of the TPC detector setup within the background limits of the radon measurement apparatus (14.0±5.7 mBq). A TPC detector run with the gas system employing 3 angstrom and 4 angstrom type molecular sieves significantly reduced the impact of common pollutants suppressing signal amplification, with the detector signal remaining until detector operation was terminated after 340 hours. Without the gas system, the TPC detector could only maintain this level of signal amplification for 50 hours. The results presented in this thesis successfully demonstrate the feasibility of a molecular sieve-based gas recycling system that simultaneously removes radon and common pollutants from SF6-based directional dark matter detectors.

Low energy nuclear recoils are one of the few signatures of the passage of WIMPs, fast neutrons and neutrinos though matter. The directional information encoded in the nuclear recoils provides valuable data which is otherwise inaccessible to a particle detector. Gas TPCs are one of the few technologies capable of reconstructing a low energy nuclear recoil track well enough to extract directional information. Scaling TPCs to large volumes while maintaining a low energy threshold and good position resolution is vital for these applications where the rarity of the interactions with matter can only be offset with larger target mass. This work focuses on amplification, charge collection and readout technologies which are able to achieve a low energy threshold on the order of keV in negative ion gases and which have the potential to scale to large areas. Initially the gain and energy resolution of the ThGEM device is determined in low pressure SF6. Results for the first operation of the novel MM-ThGEM amplification device in a negative ion drift gas are presented, showing that the device overcomes a number of problems encountered with the ThGEM while maintaining good gain in SF6. A resistive layer micromegas is used to achieve three dimensional reconstruction of events in combination with the MM-THGEM which is shown to be necessary to obtain overall gas gains sufficient to achieve a low energy threshold. The MMThGEM-micromegas is shown to work well in combination with the scalable Kobe NI-DAQ electronics and to be sensitive to alpha particles, x-rays, neutrons and gamma rays. The results indicate that the novel technology is a promising avenue of development towards a large directional nuclear recoil detector. A study of the feasibility of a gas TPC experiment aiming to observe the CEνNS scattering of reactor neutrinos is also presented for the first time.

The Short Baseline Neutrino (SBN) programme is an upcoming neutrino experiment situated on the Booster Neutrino Beam (BNB) at the Fermilab National Laboratory. One of the primary objectives of the SBN programme is to confirm or refute the low energy electron neutrino excess observed in previous neutrino experiments: LSND and MiniBooNE. It was postulated that this observed low energy electron neutrino excess was caused by the existence of one or more sterile neutrinos. If this is confirmed, it will alter our current understanding of physics as well as the standard model and the prescription of neutrino oscillations. To achieve this primary objective, the SBN programme will perform studies which are sensitive to electron neutrino appearances. These are carried out assuming several sterile models, in particular the 3+1 model. To undertake the physics goals of the SBN programme, three Liquid Argon Time Projection Chambers (LArTPCs) are positioned at various points along the BNB beamline. These LArTPCs are known as The Short Baseline Near Detector (SBND) (110 m), Micro Booster Neutrino Experiment (MicroBooNE) (470 m), and the Imaging Cosmic And Rare Underground Signals (ICARUS) (600 m) detector. LArTPCs provide sophisticated calorimetric and topological information to identify the energy and flavour of charged particles in neutrino interactions. For an electron neutrino excess search, it is important to reconstruct and identify the resultant electron from neutrino Charge Current (CC) events. A new framework with new methods was developed to characterise electromagnetic showers to help identify electrons from background photon showers. The new methods were then employed in an oscillated electron neutrino selection upon simulated events in the SBN detectors. The resultant event distributions were then used to perform an electron neutrino appearance sensitivity analysis using the 3 + 1 sterile model in the VALencia-Oxford-Rutherford (VALOR) neutrino oscillation fitting framework. The single-phase wire near detector of the SBN programme, SBND, is also viewed as a prototype for the upcoming Deep Underground Neutrino Experiment (DUNE) far detector. Due to the high rate of events at the location of the DUNE near detector, single-phase wire LArTPCs are not feasible. Therefore, alternative readout methods are being considered, such as a pixelated readout. To test these alternative readout methods, a research rig at the University of Sheffield has been developed.

There are many forms of evidence that point towards an unknown form of matter, known as dark matter, making up ∼85% of the mass in the universe. Many dark matter candidates have been proposed with the Weakly Interacting Massive Particle (WIMP) being among the most favoured. There are many groups around the world actively looking for WIMPs with direct, indirect and collider searches with specific interest here in annual modulation and directional searches. The DRIFT-IId detector is the world’s largest directional dark matter detector and is operational in Boulby Mine in the UK. Members of the directional community have come together to form the CYGNUS collaboration, looking towards larger detectors with better directional sensitivity. This thesis looks towards the future scale up to larger directional detectors, specifically low-pressure gas detectors. Improvements have been made to a system used to measure the radon emanation of materials, with emanation tests taken of potential components for CYGNUS detectors. Measurements have also been taken with a small scale THGEM TPC in both CF4 and SF6 gas. The results from CF4 showed the high gas gains achievable from the THGEM detector and allowed a direct measurement of the Townsend coefficients of the gas. Gains of up to 8600 ± 150 have been achieved in low pressure SF6 with a resolution of 19%, both of these figures are the highest achieved to date. The directional sensitivity of 1D readouts has been tested with initial signals of head-tail shown in a THGEM TPC in SF6. A head-tail signature is also seen in a simplified 1D DRIFT-IId readout mode. Exclusion limits from both the full and simplified DRIFT readouts have been produced from over 100 days of background data. The result of 0.16 pb from the full analysis is the lowest limit produced by any directional detector. These results show that a one-dimensional readout may be feasible for directional WIMP detection removing the need for many hundreds or thousands of read out channels required for 3D reconstruction.

Liquid argon (LAr) time projection chambers (TPC) have rapidly increased in importance as particle detectors throughout the past four decades. While much research has been completed, there are still many areas which require further development to build and operate the next generation LAr TPC experiment, such as the Deep Underground Neutrino Experiment (DUNE). These include high voltage breakdown, argon purification and purity monitoring, and vacuum ultraviolet (VUV) scintillation light measurement. Visual monitoring of high voltage breakdown is helpful in allowing assessment of the performance of high voltage component design. Thus, a system of cryogenic cameras, the first of its kind, was developed for use in a large LAr cryostat, without the need for additional electronics heating. The system functioned without problem for 50 days at cryogenic temperature, with some degradation of image quality, and provided a useful monitor for the DUNE 35-ton cryogenics systems. The system did not observe any high voltage breakdowns during the run. Further development of the concept is ongoing for future installation in other experiments. The monitoring of LAr purity using TPC data is a fundamental study for LAr TPC experiments. However, the study has not been performed for a large LAr TPC in the presence of high electronic noise. Custom software was developed and validated for the accurate reconstruction of signals in noisy TPC data. The results of the reconstruction were used to successfully measure the LAr electron lifetime with an uncertainty comparable to alternate methods of measurement. The electron lifetime of the 35-ton Phase II run is determined to be $4.12\pm0.17$~(stat.)~$\pm0.40$~(syst.)~ms. For general purpose research and development of high purity LAr as a particle detection medium, a dedicated test stand was designed, constructed, and commissioned. The system is used to test the gaseous photomultiplier (GPM) performance at cryogenic temperatures. The GPM functions with photoelectron multiplication at 77~K, at a reduced gain. Further study is required to show the detector's direct sensitivity to LAr VUV scintillation light.

There is now compelling evidence that ordinary baryonic matter only represents 15% of the matter content of the Universe. Observational results suggest that the remaining 85% may be constituted of dark matter possibly in the form of weakly interacting massive particles (WIMPs). One of the potential ways to detect these WIMPS is to look for their scattering interactions with nuclei. This is the basis of direct detection experiments. In particular, galactic signature direct detection experiments look for the characteristic properties of the WIMP signal. The current landscape of galactic signature experiments is dominated by two main types of experiments. Firstly, NaI (Tl) detectors are searching for the annual modulation of the WIMP recoil rate induced by the revolution of the Earth. Secondly, directional time projection chambers (TPCs) can reconstruct the momentum of the incoming scattering particle and determinate whether its origin is compatible with the WIMP wind. In this thesis, both types of experiments are addressed. For these rare-event searches, the performance of the detector is dictated by two linked parameters, the mass of target materials and the rate of background events. A new generation of galactic signature experiments is currently being developed. This work addresses the issue of the background levels through the use of Monte-Carlo simulations to predict the event rate associated with the different backgrounds. In the context of the COSINE experiment, these simulations investigate the neutron background in the detector and compare the associated rate to a theoretical model which proposes that neutrons may be responsible for the positive signal seen by the DAMA experiment. Otherwise, for the proposed CYGNUS experiments, these simulations are done in a way to facilitate the design effort of the collaboration and orientate the blueprints towards detectors which could potentially achieve background event rates below 1 per year. These efforts may potentially lead to the creation of background-free experiments larger than the DRIFT-IId TPC. Background-free status was achieved in DRIFT with the discovery of minority carriers in 2013. This thesis presents the current world-leading directional limit on the spin-dependent WIMP-proton cross section achieved with the DRIFT-IId detector. The recent detection of fast neutrons from the rock at the Boulby underground laboratory is also discussed. This is the first ever measurement of the concentration of radioisotopes in an underground laboratory using a TPC. This thesis is considering the impact that this new technique may have on future dark matter searches and how it may provide a new tool for neutron metrology in nuclear physics.

There exists compelling evidence that baryonic matter constitutes only 15% of the matter budget of the Universe. Results from a diverse range of experiments suggest that the remaining 85% is in the form of weakly interacting particle dark matter, with a particular class of particle, the WIMPs, being favoured on theoretical grounds. Recently, hints of a WIMP signal have appeared at low WIMP mass in several solid-state direct dark matter detectors. However, these appear to be at odds with the exclusion limits from the most sensitive detectors in the world, which employ liquid noble gases as their target media. The DRIFT experiment aims to measure not only the energy, but also the directionality of WIMP-nucleon interactions, which would provide an unambiguous signal of dark matter. The current generation of the detector, the 1 m3 negative ion time projection chamber DRIFT-IId, is currently taking data underground at the Boulby Underground Science Facility. This thesis presents work toward the next generation of the experiment, DRIFT-IIe, which is acting as a technology testbed for the planned 24 m3 DRIFT-III detector. The main background contributor, radon gas, is investigated, and reduced by a factor of 2 through a program of materials screening and substitution. Simplification of the electronics scheme is investigated, and found to be possible with no measurable reduction in directionality or background discrimination. A new gas mixing system for the DRIFT-IIe detector is designed and commissioned, which is more remotely-controllable and incorporates lower-cost components than its predecessor. Finally, a new technique for fiducialising events in the z dimension is presented and a new automated analysis of this data developed, which is shown to improve the efficiency for detecting WIMPs by up to a factor of 3:5.

The nature of dark matter remains one of the biggest questions in physics today. Weakly Interacting Massive Particles (WIMPs) are a particularly well motivated candidate for the missing matter that makes up 85% of the mass of the Universe. The most promising method for an unambiguous proof of the existence of WIMPs is via detection of the predicted directional anisotropy. The DRIFT detector at the Boulby Underground Laboratory in the UK is the world's first large scale directionally sensitive dark matter detector. This thesis presents work focussing on the ability of DRIFT to be competitive with non-directional detectors in exploring new spin-dependent WIMP interaction phase-space. Experimental efforts towards this are discussed, including the first calibration measurements of spin-dependent target gases in DRIFT, and development and implementation of an automated gas mixing system required for spin-dependent gas mixture operation. This thesis presents the first long-term study of backgrounds in DRIFT in which current limiting backgrounds are identified and studied, providing information crucial to future background reduction strategies. Developments of the WIMP analysis procedure are presented that result in an improved sensitivity to WIMP-mimicking neutron-induced nuclear recoils by a factor of 2.4. Data from the first runs with spin-dependent sensitive CS2-CF4 gas mixtures are presented with improved analysis methods. This thesis presents the first blind analysis results from a directionally sensitive dark matter detector with upper limits on the SD WIMP-proton interaction cross-section with a minimum of 0.93 pb for a 100 GeV WIMP.