Scott Oser

The Super-Kamiokande experiment is a 50 kiloton water Cherenkov detector located in the Kamioka Observatory in Japan. It is used to study neutrinos from the Sun, supernovae, cosmic rays and a neutrino beam produced at the Japan Proton Accelerator Research Complex 300 km away. The detector is composed of 11,129 inward facing 20-inch photomultiplier tubes (PMTs) that are used to detect Cherenkov light produced by charged particles through Cherenkov emission. Understanding the complex particle interactions that take place inside the detector is crucial to the success of the experiment. Super-K has reached a high level of precision in its measurements of neutrino oscillation parameters, yet the systematic uncertainties associated with the detector are still a limiting factor. It remains a question whether the geometry of the detector is still as it was when it was designed, or if with changes over time, or buoyancy effects on the positions and orientations of the PMTs, should be accounted for in all Super-K physics analyses. This thesis presents a novel, inter-disciplinary approach - a technique called photogrammetry from remote sensing - and proposes its application to particle physics as a way to calibrate water Cherenkov detectors. Specifically, the work develops the photogrammetry analysis pipeline, how 2D images of the inside of the Super-K detector may be used to perform its 3D reconstruction. The manuscript begins with an introduction and background to neutrino physics, Super-Kamiokande and next generation experiments, and the principles of photogrammetry. It delves into the procedures for camera calibration, detecting relevant features in the images, and the way to rebuild the detector geometry. A first ever photogrammetric reconstruction of the light injector column inside of the Super-K tank is presented as well as complementary studies which demonstrate the feasibility of the technique and the ability to achieve sub-cm precision. The results of this work are promising and motivate further development to finish the entire reconstruction for Super-K and implement the measurement into physics analyses. Furthermore, the thesis discusses hardware developments for the next generation Hyper-Kamiokande as well as how the photogrammetry for these detectors can be greatly improved.

View record

In the pursuit of directly detecting Dark Matter (DM), the SuperCDMS experiment, situated at the SNOLAB facility in Canada, focuses on studying two primary types of detectors for use in the main experiment: iZIP and HV detectors. In this analysis, we examine a specific type of detector known as High Voltage eV-resolution (HVeV) detectors. These detectors are smaller in size and offer superior resolution compared to HV detectors. The HVeV detector consists of high-purity silicon operating at low temperatures and uses the Neganov-Trofimov-Luke (NTL) effect to convert the signals of electron/hole pairs created by energy deposition into detectable phonon energy, which is measured using transition edge sensors on the surfaces of the detector. An integral part of studying HVeV detectors involves investigating the probabilities of Charge Trapping (CT), in which charge carriers are lost without receiving the full NTL amplification, and Impact Ionization (II), in which charge carriers dislodged from impurities in the crystal contribute additional photon energy through the NTL effect. Measurements of a prototype HVeV detector at TRIUMF reveal a CT probability of (12.0 ± 0.2)% and an II probability of (1.4 ± 0.2)%. A notable observation is the strong correlation between the intensity of the calibration LED and both the position of the spectrum peak and the detector resolution. We propose an explanation for this effect: the presence of non-quantized energy deposition due to photons that are absorbed without NTL amplification. We develop a model and provide substantial evidence to support this hypothesis. Lastly, we discuss the implications of this effect on calibration in previous SuperCDMS runs, highlighting the potential consequences it may have.

View record

The TUCAN (TRIUMF Ultra-Cold Advanced Neutron) neutron electric dipole moment (nEDM) Experiment is looking to measure the nEDM with a goal sensitivity of 1 x 10⁻²⁷ ecm. The discovery of a non-zero nEDM would help explain the matter-antimatter imbalance seen in our universe. The TUCAN nEDM experiment requires polarized ultracold neutrons to be stored in a volume with precisely known electromagnetic fields. To achieve high counting statistics, a large number of neutrons and high degree of neutron polarization are required. For this, the neutron polarization needs to be conserved during neutron transport, leading to the prerequisite of adiabatic transport. Adiabatic transport is when there are no strong gradients or zero field transitions such that the neutron polarization can stay aligned with the magnetic field along the neutrons’ path. An adiabatic transport system can be used to achieve this by superimposing additional fields to existing background fields to tune field properties.The development of the adiabatic guiding fields subsystem consists first of understanding adiabatic transport as quantified by the adiabatic parameter. The derivation of this parameter, and its relation to the resulting polarization of neutron ensembles, is explored in this work. Requirements on the minimum value of the adiabatic parameter for a desired minimum polarization have been set, which are then the main guidelines while designing the guide coils themselves.The adiabatic parameter is dependent on the magnetic field and angular gradients, and a thorough mapping of the guide region will be necessary for the final guiding fields design. Currently, many parts of the TUCAN nEDM Experiment are in the design phase, increasing the difficulty of obtaining meaningful field maps. In the meantime, a Magnetic Test Environment has been built and tested to model many of the major magnetic components. This setup was used to develop a mapping and analysis procedure in order to test prototype guiding coils. The coils tested were successfully able to create regions of controlled fields giving proof of principle. Future work should continue into the design and automation of this guide coil design method as more of the experimental region is built.

View record

The Weakly Interacting Massive Particle (WIMP) has historically been a prime candidate for dark matter due to its elegant compatibility with the Minimally Supersymmetric Standard Model (MSSM). Recent dark matter experiments have ruled out much of the GeV~TeV mass range predicted by the MSSM, however, and the newest generation of direct detection experiments, such as SuperCDMS SNOLAB, have begun to explore sub-GeV dark matter. As experiments push towards lower masses, sensitivity to eV-scale electron recoil events has become increasingly important. A variety of unexplained excesses at energy deposits of 1~100 eV have been found in many such low-mass experiments, across different detection techniques and at different excess rates. These low-energy events are not thought to be dark matter, and they must be understood and effectively removed to further improve detector energy resolution.This thesis will outline the simulation of one possible source of low-energy excess at SuperCDMS: optical photons produced from charges in uniform motion, specifically from Cherenkov radiation (CR), transition radiation (TR), and the intermediate hybrid transition-Cherenkov radiation (HR). In the latter case, the standard formulae for HR contain divergences at certain angles in transparent media, which are an artifact of their derivation and become problematic to simulate. To resolve this, we present a novel approach to normalize the divergent HR peaks, which allows HR to transition smoothly between TR and CR in a numerical simulation. We also add TR and CR as a physics process to the SuperCDMS Monte Carlo simulation package SuperSim, based on the Geant4 simulation toolkit. To verify the physics of our addition, we show some test simulations in comparison to theoretical predictions of optical radiation intensity. We also use the simulation to make some preliminary predictions on the contribution of TR and CR to the low energy background, with the expectation that more thorough analyses will be conducted in the future.

View record

The SuperCDMS experiment uses cryogenic silicon and germanium detectors to search for dark matter candidates such as WIMPs (Weakly Interacting Massive Particles) streaming through the Earth. Collisions in the silicon and germanium crystals are expected to produce phonons whose thermal signatures can be measured.This thesis first describes the integration of a new Signal Distribution Unit (SDU) to the SuperCDMS data acquisition system, which allows for synchronization of multiple detectors and electronic/mechanical noise characterization via accelerometer, antenna, and AC phase measurements.From SuperCDMS detector data it is necessary to reconstruct the energies of the particle events. This thesis explores the use of Convolutional Neural Networks (CNNs) to perform this reconstruction and finds that, although they perform well, changing the noise model breaks the model and requires the neural network to be retrained. In order to mitigate this issue, a new CNN model is proposed which includes the noise Power Spectral Density (PSD) of the data as an additional input to the CNN. While it proves to be effective as a denoising algorithm, it still fails for data with a different noise model. However, including data from multiple PSDs in the neural network training sample allows it to handle data with different types of noise while still maintaining the quality of the reconstruction. Nevertheless, neural networks trained even on multiple PSDs do not robustly handle data taken with PSDs dissimilar to those in the training sample, suggesting that CNNs may need to be retrained whenever the noise environment changes in a significant way.

View record

Observational evidence suggests that the majority of mass in the universe takes the form of non-luminous "dark matter". The Super Cryogenic Dark Matter Search (SuperCDMS) is a direct-detection dark matter experiment that searches primarily for a well-motivated dark matter candidate known as the weakly-interacting massive particle (WIMP). The experiment looks for an above-background excess of nuclear recoil events in cryogenic solid-state detectors that could be attributed to WIMP-nucleon collisions. The most recent SuperCDMS run at the Soudan underground laboratory set a world-leading limit on the spin-independent WIMP-nucleon cross section for a WIMP mass as low as ~3 GeV/c², and the next installation of the experiment at SNOLAB aims to be sensitive to WIMP masses below 1 GeV/c². To better understand the response of solid-state germanium detectors to low-mass WIMPs, "photoneutron" calibration data was taken at the Soudan laboratory in Minnesota by passing quasi-monoenergetic neutrons through SuperCDMS detectors. Gamma rays used in the photoneutron production process create an overwhelmingly dominant background of electron recoil events in the detector. This gamma background is measured directly with regular "neutron-off" data-taking periods during which the neutron production mechanism is removed. We compare the observed electron and nuclear recoil spectra with Geant4-simulated spectra to obtain a model-dependent calibration of the nuclear recoil energy scale of the detectors. The calibration is performed using a negative log likelihood fit to a parameterized Lindhard ionization yield model. The fit includes a semi-analytical model of the gamma background component obtained from the neutron-off data.

View record

Long-baseline neutrino experiments produce neutrinos by colliding a beam onto atarget. Hadrons produced in the target can be magnetically focused into a beam thatdecays in flight, yielding a collimated beam of neutrinos. Neutrino flux predictionsare used to constrain the uncertainties of the detected neutrino flux and its uncertaintyis dominated by the uncertainties in hadronic interactions. The reductionof this uncertainty would further the physics goals for these experiments. Theseuncertainties can be sourced to the production cross section uncertainty which hasbeen set to a conservative value due to previous discrepancies in the estimated productioncross section seen in the NA61/SHINE 2cm and 90cm carbon target data,used by the T2K collaboration. In addition, untuned secondary interactions withinthe horn and decay volume contribute to the uncertainty.The EMPHATIC group has taken data using the Fermilab test beam on carbon,aluminum, and steel targets for hadrons ranging from 2GeV/c to 120GeV/c.EMPHATIC uses gas Cherenkov and lead glass calorimeters for particle identification,semiconductor tracking detectors for position and timing information, andemulsion situated on a moving table for precise position measurements. In addition,an aerogel Cherenkov detector was constructed at TRIUMF which was usedfor testing at the Fermilab Test Beam Facility. The silicon strip tracking detectorswere instrumental in the data acquisition of the EMPHATIC experiment. The dataacquired by the silicon strip detector for 30GeV/c protons impinging on a carbontarget was fit with the Bellettini et al. model . The parameters of the total cross section, the black body radius, and the nuclear transparency term were fit and agreewithin one standard deviation with previous measurements of Bellettini et al. Theelastic cross section agrees within two standard deviations.

View record

The SuperCDMS SNOLAB experiment will use solid state Germanium and Silicon detectors to search for Weakly Interacting Massive Particles (WIMPs), a leading candidate to explain dark matter. WIMPs are thought to exist in halos around galaxies and therefore thought to be constantly streaming through the earth. The CDMS detectors have been developed to measure the energy deposited by a WIMP-nucleon collision in terrestrial calorimeters.This thesis focusses on the Data Acquisition (DAQ) system that uses Detector Control and Readout Cards (DCRCs) and is designed to be dead- time-less. The DCRCs read in the data stream from the detector’s 12 phonon and 4 ionization energy channels. The DCRCs also control detector settings, and we develop interactive codes to allow users to easily change detector settings through the DCRC. The DAQ is designed to decide which events to write to disk in order to keep data throughput under a limit yet never miss an event that will be useful in the subsequent analysis. In this effort we develop different readout methods of the detector array for the different calibration runs and WIMP search runs. We also develop fast algorithms for rejecting events that fail a certain criteria for being usable. We also present a novel data compression method that reduces the total data volume by a factor of ∼ 16 yet retains all important information. This method involves a large covariance matrix inversion, and we show that this inversion can be consistently computed given that a sufficient amount of data has been used to build the covariance matrix. We also develop a GUI that is a critical element of the detector testing program for SuperCDMS SNOLAB. The GUI accesses the data stream as it is being written to disk, efficiently reads in the waveforms, and displays them in a user-friendly, oscilloscope-like, format. By making use of Fast Fourier Transform technology, the GUI is also capable of displaying incoming data in the frequency domain. This tool will enable a new degree of real-time analysis of detector performance, specifically noise characteristics, at the test facilities in the next stage of detector testing.

View record

The T2K experiment is a long baseline neutrino experiment. Rates of different types of neutrino interactions are compared at a near detector and a detector 295 km from the neutrino source which enables neutrino oscillation and cross-section measurements. Understanding the activity near the neutrino interaction vertex in the near detector is crucial to identifying how a neutrino interacted. Current reconstruction algorithms for T2K's Fine Grained Detectors (FGD) have a decreased sensitivity to short tracks (200mm) such as those created by low energy protons and pions. A new likelihood based reconstruction algorithm entitled FLikFit (FGD Likelihood Fitter) has been written that uses information from the deposited charge and the position of the hits within the FGDs in an attempt to tag these particles. It traces a predicted path of the particle through the detector and calculates expected kinetic energy and path length values which are fed to an existing probability distribution function to obtain a likelihood for each scintillator bar. The fitter is seen to calculate likelihoods effectively and proves to be adept at hypothesis testing, but further development is required to improve the FLikFit's ability to converge.

View record