School of Physics - Theses
Permanent URI for this collection
Search Results
1 - 10 of 243
-
ItemFinding Quasars in the Southern Hemisphere Sky Using Random Forest Machine Learning( 2022-04)Quasars are the most luminous persistent sources in the universe. The main goal of this thesis is to search for quasars at different redshifts using an efficient machine learning algorithm: the Random Forest classifier. A technique was developed and tested on a small photometric sample, the Early Data Release (EDR) from the SkyMapper Southern Survey (SMSS). After classification by Random Forest, candidates were prioritised for confirmation using observations from the ANU 2.3m telescope. SkyMapper is the first digital optical survey in southern hemisphere and has been used to build the required training subsets and the dataset. SMSS has been matched with other available surveys in the southern hemisphere to provide a broader range of colours for selection algorithms. The predictions were greatly improved by combining photometric colours in the optical from SMSS with mid-infrared data from AllWISE. Random Forest Machine Learning techniques provided classifications with probabilities of up to 81%. The EDR pilot study, predicted 119 QSO-candidates. Of these 78 have been confirmed as quasars, either previously or by new observations, and the remainder still need to be observed. So far, only one galaxy and one star were found amongst the candidate list. In addition, the classifier has been trained and applied to a much larger dataset, the Third Data Release of SMSS. This provides a preliminary study of the techniques that will be required to study extremely large samples of quasars identified in the Legacy Survey of Space and Time which will commence in a few years time on the Vera C. Rubin Observatory. Other observations in the thesis explored different AGN types, such as narrow line objects, using the BPT diagnostic diagram to determine the source of excitation energy. Future machine learning algorithms may be able to determine finer AGN classifications, as the range and quality of the input non-spectroscopic datasets improves. The conclusion from these studies is that candidate quasars can be identified with high confidence using machine learning, if a sufficiently large spectroscopic test sample is available.
-
ItemSimulating Noisy Quantum Algorithms and Low Depth Quantum State Preparation using Matrix Product States( 2021)Since the proposal of Quantum Computation in the 1980s, many Quantum Algorithms have been proposed to solve problems in a wide variety of fields. However, due to the limitations of existing quantum devices, analysing the performance of these algorithms in a controlled manner must be performed classically. The leading technique to simulate quantum computers classically is based on the Matrix Product State (MPS) representation of quantum systems. We used this simulation method to benchmark the noise tolerance of a number of quantum algorithms including Grover’s Algorithm, finding that the algorithm’s ability to discern the marked state is exponentially suppressed under noise. We verified the existence of Noise-Induced Barren Plateaus (NIBPs) in the Quantum Approximate Optimisation Algorithm (QAOA) and found that the recursive QAOA (RQAOA) variation is resilient to NIBPs, a novel result. Also integral to the performance of quantum algorithms is the ability to efficiently prepare their initial states. We developed novel techniques to prepare low-depth circuits for slightly entangled quantum states using MPS. We found that we can reproduce Gaussian and W States with circuits of O(log(n)) depth, improving on current best known results which are of O(n).
-
ItemDetector characterisation for SABRE South: A NaI(Tl) based dark matter direct detection experiment( 2023-03)The existence of a dark, non-baryonic matter component of the universe is a core component of modern astrophysical models, yet decades of experimental searches are yet to observe non-gravitational dark matter interactions with Standard Model particles. The DAMA experiments have detected a modulating signal compatible with dark matter for over twenty years, with a combined significance of 12.9 sigma. This result is in tension for a spin independent WIMP with the null results from other experiments. Sodium iodide with Active Background Rejection (SABRE) South is the Southern Hemisphere's first dark matter direct detection experiment, designed as an independent replication of the DAMA modulation search using the same NaI(Tl) target, and due to begin construction this year. SABRE South consists of an array of seven ultra-low background NaI(Tl) crystals, each coupled to two high efficiency R11065 photomultipliers. These are located within an active veto comprised of 12,000 litres of liquid scintillator instrumented with eighteen oil-proof R5912 photomultipliers. This thesis will focus on the characterisation of several key detector components for SABRE South, detailing the requirements of the SABRE South sub-detectors necessary to achieve the physics goals of the experiment. This thesis describes the photomultiplier characterisation methods and analysis for SABRE South, these measurements include the single photoelectron response, the rate of dark events as a function of temperature and voltage, the quantum efficiency, and studies of the timing response. Results are reported for two R11065 photomultipliers used to establish and validate the system, alongside preliminary results from the bulk characterisation of the twenty R5912 photomultipliers. The characterisation of each individual photomultiplier is critical to achieving the low energy sensitivity required by SABRE South. This thesis also describes the measurement and analysis of the first ultra-pure SABRE South NaI(Tl) prototype crystal, NaI-35, which was performed underground at the Gran Sasso National Laboratory. The crystal light yield was measured to be 9.29 +/- 0.14 photoelectrons per keV at 59.5 keV. The energy resolution was measured at 59.5 keV as 5.9 +/- 0.5% with further measurements at a range of energies between 20 and 300 keV used to fit a model of energy resolution. The rate of intrinsic alpha decays in the crystal was measured by direct counting as 0.529 +/- 0.013 mBq/kg which is less than half that reported by ANAIS and COSINE. This can be used to constrain the activity of 210Pb, the dominant background in SABRE South. The SABRE digitisation software package used to emulate the detector for Monte Carlo studies is introduced, discussing the method used, and the validation against a real detector. This software is essential to the development of accurate signal classifiers, and to quantify the effect of photomultiplier noise on reconstruction and the experiment background. The accurate characterisation of the photomultipliers and NaI(Tl) crystals, and the development of tools to enable their accurate replication in Monte Carlo simulations is crucial to ensuring that the SABRE South experiment has sufficient sensitivity to not only replicate the measurement of DAMA, but to study any potential signal providing clear evidence on its origin and nature.
-
ItemThe characterisation and implementation of nitrogen-doped ultrananocrystalline diamond photoelectrodes for biological applications( 2023-01)The development of electrodes for the electrical modulation of cell activity has become a research topic of increasing interest, due to an ever-growing list of applications. These include deep brain stimulation for neurodegenerative disorders such as Parkinson’s and Alzheimer’s disease; implantable bionic devices such as the Cochlear ear implant and bionic eye; and promoting cell growth for wound healing or tissue engineering. Electrical stimulation is conventionally achieved by delivering current from an external power source through metallic electrodes to the target biological tissue. However, these types of electrodes have drawbacks. For instance, they typically require cumbersome and invasive wiring, while the stimulating spatial resolution is limited by current spreading. One attractive alternative technique is light-based stimulation, which offers wireless and scalable targeting of cells with superior spatial precision. In particular, light-sensitive electrode materials can be used to transduce light into an electrical stimulus to excite nearby cells. Nitrogen-doped ultrananocrystalline diamond (N-UNCD) is a promising material for this technique, possessing a highly attractive combination of properties including high chemical inertness, durability, conductivity, and biocompatibility. In addition, it has also been shown to produce a photoresponse using near-infrared light, which offers greater optical penetration in biological tissue than shorter wavelengths. In this thesis, we optimise and characterise N-UNCD electrodes for light-based cell stimulation, and then demonstrate their use in the modulation of stem cell cultures. To achieve this, we tested the effect of different chemical surface treatments on the stimulating performance of the N-UNCD electrodes, measured by the capacitance and photoresponse. Of these treatment methods, we found that annealing N-UNCD in oxygen ambient produced dramatically improved properties, with the capacitance and photoresponse reaching values of 28.5 +/- 0.3 mF cm-2 and 3.75 +/- 0.05 uA W-1, respectively. This translates to an enhancement of up to 17 times compared with the previously used oxygen plasma treatment. We then investigated the reason for this enhancement, finding that it was due to a combination of factors. These included conductive grain boundary etching, and the chemical functional groups on the electrode surface changing the alignment of band-edge positions with respect to energy levels of redox species in solution. These properties were incorporated into a compact computational model which provides an overarching description of the photoelectrochemical data from the N-UNCD electrodes, from experiments including cyclic voltammetry, electrochemical impedance spectroscopy, and photocurrent measurements. Following this, we conducted a theoretical evaluation of the stimulating efficacy of the optimised electrodes, concluding that the photocurrent dynamics are best suited for long-term photomodulation applications rather than the short-term high intensity stimulation of neurons. Therefore, we decided to test the electrodes for long-term light-based stimulation of human mesenchymal stem cells (hMSCs). We found that the N-UNCD material itself promotes the process of osteogenesis (bone cell development) in the stem cells, which may be due to its extremely hard and nanotextured surface. Moreover, initial short term (6 hours) pulsed illumination of hMSCs cultured on the N-UNCD electrodes produced significant long term increases in cell proliferation and differentiation over 21 days. These results may assist in the development of stem cell techniques for tissue engineering, including for orthopaedic implants and bone healing therapies.
-
ItemDesigning and assessing model independent tests of the DAMA modulation( 2022-12)Particulate dark matter is a long hypothesised solution to various astrophysical observations seemingly at odds with a completely luminous universe. Despite the success of dark matter in explaining these observations, to date physicists have been unable to conclusively observe its interactions with Standard Model matter directly. This thesis will focus on trying to understand the results from the DAMA collaboration, which for the past two decades has reported a modulation signal consistent with dark matter, but in tension with other null experimental results under the usual dark matter assumptions. This study demonstrates the need for a model independent test of this signal to understand its origin, the requirements of such a test, and how different dark matter experiments can be compared or assessed to understand how sensitive they are to this elusive signal. This thesis examines such a study through the lens of a dark matter detector currently under construction in Australia: SABRE South. In particular, it will focus on purification techniques that can be used to produce benchmark low background equipment, detailed simulation studies that can guide the design of SABRE South, and the detailed analysis that must take place to understand how sensitive and or competitive SABRE South will ultimately be. It will also touch on interesting phenomenology studies that can be conducted with such a detector; examining non-standard or unusual dark matter models and signatures that are produced by relaxing the assumptions typically made about its fundamental nature, and distribution with the galaxy.
-
ItemSemi-leptonic decays at the Belle and Belle II experiments( 2022)This thesis explores semi-leptonic decays of the type $B \to V L$ along two avenues; their contribution to high-precision tests and metrology of the Standard Model, and their potential for probing physics beyond the Standard Model. The magnitude of the Cabbibo-Kobayashi-Maskawa matrix element corresponding to $b\to c$ quark transitions, $|V_{cb}|$, has a long-standing discrepancy between its measurements from exclusive and inclusive decays, which presents a puzzling gap in our knowledge of the Standard Model. In this thesis, recent Lattice QCD calculations of form factors at nonzero hadronic recoil are incorporated into novel analysis methods to examine the phenomenology and analysis techniques behind the exclusive $|V_{cb}|$ measurement in $B \to D^{*}\ell^+ \nu_\ell$ decays. This results in a measurement of $|V_{cb}|=(38.49 \pm 0.54 \pm 0.92 \pm 0.91) \times 10^{-3}$, the least model-dependent result to date. The results of this analysis of the $B \to D^{*}\ell^+ \nu_\ell$ branching fraction and decay rate are used in a test of Standard Model QCD predictions of hadronic factorisation, with the first measurement of the parameter $|a_1|$ within a single experiment, based on a ratio that cancels most experimental systematic uncertainties, resulting in $|a_1| = 0.884 \pm 0.004 \pm 0.003 \pm 0.016$. This corresponds to an $8.9\sigma$ discrepancy from the Standard Model, suggesting the existence of large non-factorisable contributions in hadronic $B$-decays, or potential contributions from new physics amplitudes. A probe for new physics in photon-dipole interactions, described by the $\mathcal{O}_7$ term in an operator product expansion, is then performed to constrain non-Standard Model right-handed currents in flavour-changing neutral current semi-leptonic decays. New lepton identification techniques in the Belle experiment are developed for use in the first study of $B \to K^* e^+ e^-$ decays at the very low dilepton invariant mass region, $q^2 < 1.12 \gevccsq$, to constrain the Wilson coefficient $\mathcal{C}_7$, and its right-handed counterpart $\mathcal{C}_7^{\prime}$, expected to be zero in the Standard Model.
-
ItemInvestigating a Third Parameter in the Tully-Fisher Relation( 2022)The scaling relations of galaxies play an important role in constraining the formation and evolution of galaxies. One of the key relations, the Tully-Fisher relation (TFR) also has a crucial place in determining galaxy distances. The relation states that absolute magnitudes of spiral galaxies are correlated with their maximum rotational velocities. This thesis focuses on how the TFR can be improved by investigating the use of an additional parameter. Following the Fundamental Plane, central velocity dispersion is studied as this parameter. Additionally, with the help of hydrodynamical simulations, possible biases are explored in the determination of the TFR. The optical data of the thesis is obtained using the WiFeS (Wide Field Spectrograph) on the ANU (Australian National University) 2.3m Telescope at Siding Spring Observatory. The sample was chosen from the HICat (HIPASS (HI Parkes All-Sky Survey) Catalogue). As velocity dispersion is the main focus of the thesis, early type spiral galaxies (S0 to Sc types) are preferred because of their velocity dispersion supported bulges. In total 137 HIPASS galaxies were observed during total of 32 nights of observing over the course of 2.5 years. The related HI data is taken from the HICat. Obtaining both HI and optical rotational velocities allowed comparison of the TFR with K band magnitudes at both wavelengths. The results are in agreement with the literature where HI data gives less scatter and a tighter correlation in the TFR. Furthermore, morphology and inclination are also studied. It is found that morphological type is not a cause for the scatter in the relation whereas there is a clear indication that higher inclination angles correlate better than the smaller inclinations. With the velocity dispersion that is obtained from the optical data, the $\sigma/V_c$ parameter is investigated as a possible third parameter. Even though adding this parameter does not result in different TFRs, there is a notably greater scatter for the earlier type of spirals, where $\sigma/V_c > 0.6$. Lastly, the $S_K$ parameter, which also consists of $\sigma$ and $V_c$, is studied to determine whether the correlation has less scatter. It is found that introducing $S_{0.5}$ reduces the scatter as in the earlier studies. Observational data is then compared with the simulated data of the EAGLE (Evolution and Assembly of Galaxies and their Environments) project. The Recal-L025N0752 model at $z=0$ is used to obtain the data. All the galaxies with $M_{\star} > 10^{10} M_{\odot}$ within a radius of 30kpc are extracted and total of 81 galaxy data is obtained to study the TFR. As the simulation has the velocity and mass of the star and gas particles, velocity dispersion and rotational velocity widths are also calculated for each galaxy. However, to mimic the observations, velocity dispersions are calculated randomly from a similar ratio of apertures as the optical sample. It is found that instead of aperture of the galaxy from which velocity dispersion is calculated, the selection of the individual galaxies affects the slope of the TFR. Using the $S_{0.5}$ parameter does not make a difference to the result.
-
ItemThe Panoramic Deep Fields( 2001)The Panoramic Deep Fields are a deep multicolour survey of two ~ 25 ° fields at high galactic latitude. The survey images have been constructed by digitally stacking scans UK Schmidt plates. Deep images (Bj ~23.5) with low contamination have been obtained by subtracting the background from the individual plates scans and using bad pixel rejection during the stacking. The size and depth of the fields allow the accurate statistical measurement of the environments and evolution of galaxies and AGN. The clustering of galaxies and galaxy clusters has been measured from z ~0.4 until the current epoch. The clustering properties of galaxies are strongly correlated with colour and blue U – Bj selected galaxies exhibit weaker clustering than any morphologically selected sample. The weak clustering (ro ≤ 3h -1 Mpc) of blue galaxies implies galaxy colour and stellar population are more strongly correlated with environment than galaxy morphology. Despite the large fields-of-view, the clustering of red galaxies and clusters varies significantly between the two fields and the distribution of clusters is consistent with this being due to large-scale-structure at z ~0.4. The evolution and environments of AGN have been measured at intermediate redshifts with the Panoramic Deep Fields. Photometric redshifts, colour selection and the NVSS have been used to compile a catalogue of ~ 180 0.10 < z< 0.55 radio galaxies. The evolution of the radio galaxy luminosity function is consistent with luminosity evolution parameterised by L (z) ~ L(0) (1+z)3.4. The environments of UBR selected AGN and radio galaxies have been measured at z~0.5 using the Panoramic Deep Field galaxy catalogue. By applying photometric red-shifts and colour selection criteria to the galaxy catalogue, it has been possible to increase the signal-to-noise of the angular correlation function and measure the cross-correlation with specific galaxy types. Most AGN host environments are comparable to the environments of galaxies with the same morphology. However, ~6% of UBR selected AGN are in significantly richer environments. No significant correlation between AGN luminosity and environment was detected in the Panoramic Deep Fields. The richness of AGN environments is not strongly correlated with redshift and the rapid evolution of the AGN luminosity function is not caused by evolution of AGN host environments.
-
ItemCharacteristic X-ray Spectroscopy( 2022)X-ray spectroscopy is the study of radiation emitted as a result of electronic transitions inside an atom. X-ray spectroscopy provides otherwise unobtainable information regarding the structure of the atom and has been vital in our understanding of the quantum mechanical laws that are now fundamental to modern science. The accuracy of recent Cu Kalpha measurements have reached the one part per million level. However, discrepancies between measurements have been revealed. A new measurement of the Cu Kalpha spectrum, using a novel detection system, allows the investigation of these discrepancies as well as satellite contributions and the relativistic quantum mechanics that underpin them. Our new characterisation is one of only three in the literature that allows the relative intensity of the Kalpha3,4 satellite to be measured. Our characterisation implies a 2p shake probability of 0.5%. In addition, this research highlights critical questions regarding the portability and transferability of x-ray spectra and helps consolidate the Cu Kalpha peak energy as a standard laboratory measurement. Contributions from the radiative Auger effect (RAE) have often been unmodelled in characterisations of x-ray spectra, resulting in incorrect conclusions being drawn from empirical fits. For the first time, the Cu Kbeta spectrum has been characterised including a component to model the RAE. The results show our characterisation to be superior to the traditional 5-Lorentzian model and also explains long-standing residuals of empirical fits around the low energy Kbeta1,3 tail. An observed and imputed fitted Auger contribution of 2% demonstrates the need for Auger components to always be considered when characterising x-ray spectra and highlights the importance of developing new atomic theory to model the radiative Auger effect using an ab initio method. Theoretical predictions of atomic spectra have been pivotal in continuing our understanding of fundamental physics. The multiconfiguration Dirac-Hartree- Fock (MCDF) method represents our best avenue for approximating the wave function of multi-electron atoms and molecules. However, the open shells of the 3d transitions metals provide significant challenges for convergence. Using the MCDF method, we present the most comprehensive calculation of transition probabilities and energy eigenvalues relating to Cu Kalpha and Cu Kbeta. Initial and final state wavefunctions show convergence, with the expansion of the active set, to 0.05 eV or 0.0006 %. The methods presented in this thesis for obtaining complete, well-converged wavefunctions for multi-hole initial and final states overcame numerous challenges and provide an approach for other researchers in computing wavefunctions for any complex system. Although we have shown that the multiconfigurational framework is necessary and capable of obtaining well-defined wavefunctions of open-shell systems, the calculation of shake probabilities using these multiconfiguration wavefunctions have continued to struggle to reproduce experimental results. This thesis provides the most complete calculation of shake probabilities using MCDF wavefunctions and the sudden approximation, while also addressing incorrect assumptions about their implementation when matching experiment. These calculations bring us a step closer to predicting x-ray emission spectra using a fully ab initio method. The significant discrepancy between satellite intensities measured experimentally and ab initio calculations of the shake processes that lead to these satellites, represents a significant challenge in the field. We test our MCDF calculations and ab initio shake probabilities by using them to simulate x-ray spectra and fitting to the best available experimental data. For the first time the Kalpha1,2 spectrum is modelled using the diagram lines and four spectator vacancy satellites: the 3d9, the 4s0, the 3d94s0 and the 3d8. We show that fixing satellite intensities to ab initio shake probabilities can simulate peak energies to within 0.1 eV of experiment. Furthermore, our MCDF calculations are used to model the Kalpha3,4 feature with agreement between theory and experiment far greater than ever before.
-
ItemConstraining cosmology with SPT-3G( 2022)Cosmic Microwave Background (CMB) observations are rich in cosmological information and a key pillar of modern cosmology. The South Pole Telescope offers a particularly clear view of the millimetre sky and its latest receiver, SPT-3G, facilitates unprecedented measurements of CMB anisotropies on small angular scales. In this thesis, we present measurements of the CMB temperature and polarisation power spectra using SPT-3G data recorded in 2018 covering a region of approximately 1500deg^2 in the southern sky. We also present the first comparison of covariance conditioning schemes for CMB data at the matrix- and the parameter-level. For empirical covariance estimators using less than or approximately 100 independent data realisations, moderate conditioning schemes can lead to an underestimation of the parameter error by up to a factor of 1.3 compared to the uncertainty inferred from the likelihood. These results inform the analysis SPT-3G data. We demonstrate the internal consistency of the SPT-3G 2018 temperature and polarisation band powers across frequency bands and spectra, and find excellent agreement. We present cosmological constraints using the polarisation spectra and report consistency with the standard model. We find H0 = 68.8 km/s/Mpc, which is considerably lower than the most precise local determination of the expansion rate. Cosmological constraints from SPT-3G and Planck data are in good agreement. We produce constraints using SPT-3G 2018 data alone and in combination with other data on a series of model extensions drawn from the following parameters: the effective number of neutrino species; the primordial helium abundance; the sum of neutrino masses; the mass of a sterile neutrino; the mean spatial curvature; the baryon clumping induced by primordial magnetic fields; the initial field value, critical redshift, and contribution to the energy density at the critical redshift of an early dark energy component. We discuss the constraints on each model in detail and report no statistically significant and robust preference for any of these extensions over the standard model.