Theses - Physics

Permanent URI for this collection

https://www.repository.cam.ac.uk/handle/1810/214775

Browse

Now showing 1 - 20 of 583

Open Access
Machine learning approach to model the microstructure and strength of nickel superalloys
Taylor, Patrick; Taylor, Patrick [0000-0001-8529-0657]
Nickel superalloys are a class of materials that find crucial applications in technologies such as jet and gas turbine engines. In order to accelerate the further development of these alloys, this thesis develops machine learning models that can better predict their microstructure and strength. The Gaussian process regression (GPR) models of microstructure are shown to be just as good at interpolation as traditional CALPHAD models, with advantages in speed, retrainability, incorporation of non-equilibrium effects, and the effective inclusion of computational data. By incorporating domain knowledge, it is shown that such GPR models can extrapolate into regions of composition space which include elements unseen during the training process. They can also make accurate predictions for the evolution of microstructure when heat treatments are applied. By making use of domain knowledge, similar extrapolations are possible for models of creep strength. Incorporating the results of the microstructure models leads to models of creep strength that meaningfully reveal underlying physical mechanisms of creep deformation.
Open Access
Numerical modelling and design of wideband electromagnetic structures at radio frequencies: Applications in cosmology and digital communications
Cumner, John
This thesis discusses the use of computational electromagnetic simulation technology to simulate various physical scenarios for analysis and design. Digital communications continue to dominate modern communication, and require increasing bandwidth to transfer the volume of information in use in today's society. One method for allowing these expansions is to improve the transmission through currently existing infrastructure, through improved impedance matching. The use of the surface wave mode of transmission is also considered to help improve data throughput, with a focus on the impedance of the required launchers. Simulated measurements for these impedances are calculated, with reference to their physical origins. Also examined within this thesis is the application of computational electromagnetics to the global 21 cm experiment REACH. This experiment aims to detect a signal five orders of magnitude below the foreground signal, and so requires highly accurate and precise understanding of the radiometer instrument. Both the design and analysis of aspects of the REACH dipole radiometer are considered throughout this thesis. While simulation of physical scenarios is to some degree accurate, it is inevitable that uncertainties will arise from simplifications made in computational models. So in the context of a dipole antenna's directivity pattern, and corresponding antenna temperature, the use of a parameterized sum of basis functions is considered to remodel the directivity of a physically perturbed antenna. Through the use of physically based basis functions and per frequency fitting, a rebuild accuracy within 0.1% of directivity is shown. The effect of likely physical deviations in a ground plane is considered in the case of the Radio Experiment for the Analysis of Cosmic Hydrogen (REACH) dipole radiometer. The physical deviations considered include the presence of soil, multiple and single dips in the ground plane. These alterations often induce uncertainties above 1 K, a level which would obscure a global 21 cm signal detection. Also noted is the impact of the addition of serrations to the edge of a square ground plane, with the replacement of large wavelike chromatic fluctuations with smaller pockmark type deviations. Finally, I describe the implementation of a quantified figure of merit based design method of radio frequency electromagnetic situations. This is used for the design of the REACH global 21 cm dipole radiometer. The figures of merit considered encompass the impedance of the antenna in addition to the chromaticity of its directivity pattern and are combined in such a way as to allow even comparison between these important aspects of the antenna.
Embargo
Ultrafast spatiotemporal dynamics of photoexcitations in two-dimensional semiconductors
Gauriot, Nicolas
This thesis focuses on two families of Van der Waals materials: the transition metal dichalcogenides and the group IV monochalcogenides. First, we explore the spatio-temporal dynamics of photo-excited species in monolayer transition metal dichalcogenides. Using pump-probe spectroscopy, and a range of steady state characterization techniques (ellipsometry, reflectance, photoluminescence), we study the formation and decay of excitons in these materials at low excitation densities. We find that the photodynamics are largely dominated by the screening of the Coulomb interaction by excited states. At high excitation densities, when the screening of the Coulomb interaction is such that the excitons are no longer stable, we observe their complete ionization into a dense plasma of unbound electrons and holes. This interaction driven transition from an insulator a metallic state is a typical Mott transition. Combining pump-probe techniques with microscopic imaging allows us to observe these processes in space with 10 fs temporal resolution and 10 nm spatial localisation. Above the Mott transition, we observe an ultrafast spatial propagation of the excitations in the first few hundred femtoseconds after excitation. We hypothesise that this new regime of ultrafast transport is driven by the Fermi pressure of the dense electron-hole gas. Because atomically thin materials are very sensitive to their environment, it is also possible to tune the energies of electronic states through changes in the surroundings. Here, we engineer a potential landscape for excitons in the monolayer by introducing spatial variations in the dielectric environment. We can then watch excitations evolve in this landscape on an ultrafast time scale. Specifically, we observe the funnelling of excitons in a lateral homojunction. Finally we demonstrate the exfoliation down to the monolayer of two representative member of the group-IV monochalcogenides family.
Open Access
Beauty meson to double open charm decays with the LHCb detector
Bishop, Fionn
Measurements of beauty meson to double open charm decays are presented. These measurements use data collected in proton-proton collisions by the LHCb detector at the Large Hadron Collider (LHC) between 2011 and 2018. Firstly a search for sixteen rare decays of the $B_c^+$ meson to two charm mesons is performed. Studies of $B_c^+$ meson can improve understanding of heavy quark dynamics, and its decay to two charm mesons could be useful in measuring the Cabibbo-Kobayashi-Maskawa quark-mixing phase $\gamma$ in future LHC data taking periods. The first evidence of the $B_c^+\rightarrow D_s^+ \overbar{D}^0$ decay is found with a significance of 3.4 standard deviations and its branching fraction is measured. Upper confidence limits on the branching fractions of the other fifteen decays are improved. Secondly, the charge-parity asymmetries ($\mathcal{A}^{CP}$) and branching fractions of decays of the $B^-$ meson to two charm mesons are measured. No evidence of charge-parity violation is seen, but the first measurements of $\mathcal{A}^{CP}(B^- \rightarrow D^{*-}_{s}D^0)$ and $\mathcal{A}^{CP}(B^- \rightarrow D^-_sD^{*0}$ are made and the precision is improved on the $\mathcal{A}^{CP}$ of a further five decay modes. Moreover, ratios between the branching fractions of doubly charmed decays of the $B^-$ meson are measured for the first time. These results will help to constrain physics beyond the Standard Model. Finally, a generic beauty hadron neural network and exclusive selections for doubly charmed beauty meson decays are developed, which are implemented in the trigger for future data taking. The improvement in selection efficiency using these selections in the upgraded trigger system is evaluated. Additionally, prospects for measurements of doubly charmed beauty meson decays with future LHCb datasets are discussed. Sensitivity to rare doubly charmed $B_c^+$ meson decays and high-precision measurements of charge-parity violation will be enabled by the upgraded LHCb detector and trigger system.
Embargo
Accelerating Materials Discovery for Optical Applications using Machine Learning, Natural Language Processing and Density Functional Theory
Zhao, Jiuyang; Zhao, Jiuyang [0000-0002-2561-352X]
This thesis presents a novel approach that combines materials informatics and theoretical calculations to accelerate the discovery of materials with desirable optical properties. Unlike traditional experimental research programmes, this work emphasises the significant contributions in terms of knowledge and technological outcomes resulting from the study. The thesis begins by reviewing the fundamental physics of optical properties of materials and the existing literature on materials informatics and the applications of machine learning in materials discovery (Chapter 1). Subsequently, the methodologies employed throughout the research are outlined, including the utilisation of natural language processing (NLP) tools for information extraction, various machine learning models, and techniques for quantifying π-conjugation in organic molecules (Chapter 2). The research presents compelling results, starting with the development of a complete workflow that extracts refractive indices and dielectric constants from scientific publications, resulting in a substantial database comprising 109,880 records of experimental data on optical materials (Chapter 3). Building upon this, second-order Sellmeier equations are used to reconstruct chromatic-dispersion relations of various compounds, while machine learning techniques are employed to model refractive indices of inorganic compounds, showcasing the potential of auto-generated databases for property prediction and visualisation (Chapter 4). Moreover, a new algorithm or metric is introduced to characterise π-conjugation in organic molecules, which plays a crucial role in determining their nonlinear optical properties (Chapter 5). This algorithm enables a high-throughput computational study on more than 20,000 molecules, leading to the identification of four commercially available organic molecules that hold sufficient potential to be used as nonlinear optical materials (Chapter 6). This metric enables the accelerated discovery of organic compounds with exceptional molecular hyperpolarisability coefficients, a domain where literature data is notably scarce. To enhance data extraction capabilities in the field of optical materials, two novel neural network-based language models, OpticalBERT and OpticalTableSQA, are presented (Chapter 7). OpticalBERT is a BERT-based model that is pre-trained on an extensive corpus of optical materials and exhibits remarkable advancements in various NLP tasks compared to traditional rule-based approaches. OpticalTableSQA is a table-based question-answering model that is specifically designed for question-answering tasks on optical materials tables, further enhancing data extraction capabilities. This thesis concludes by summarising the contributions made and outlining potential avenues for future research (Chapter 8). By leveraging materials informatics, machine learning, and novel metrics for organic compounds, this research significantly advances the discovery and understanding of materials with desirable optical properties. The developed methodologies and databases provide valuable resources for researchers in the field and pave the way for further exploration in this domain.
Open Access
Modelling of mechanosensitive morphogenesis and maintenance of cell structure: Application to musculoskeletal cells and tissues
Ibata, Neil
My dissertation presents some of the first key steps towards an analytical understanding of mechanosensitive morphogenesis and structure maintenance in biological systems. In order to make this problem tractable and to make a significant advance during the PhD, I focussed specifically on musculoskeletal applications of mechanosensing. My approach to this problem was to break it into two pieces: nucleation of structure, and how external and internal loads maintain and adapt existing structures. From a physical perspective, the first of these involves symmetry-breaking and can be studied with the tools of phase transitions, whereas the second involves out-of-equilibrium dynamics and requires a numerical modelling approach of time-dependent non-linear differential equations. **Nucleation of clusters of adhesive molecules.** In order to investigate the nucleation of structure under load, I considered the initial nucleation of the adhesive molecules integrin and cadherin. These are of paramount importance in holding animal tissues together, because they allow cells to bind either to the extracellular matrix or to other cells. Both of these molecules form similar chains of molecules that strengthen under load and bind to backbone of the cell, the cytoskeleton. By doing so, they help to develop many of the structures in the body: integrin-rich fibroblasts lay out collagen fibres in tendon and bone, whilst cadherins hold epithelial soft tissues as well as cardiac muscle together. The nucleation of clusters of adhesive molecules is the first step towards specialising the cells that contain them in order to develop and strengthen many of the structures in the body. It has recently been shown that they can aggregate if they can display some binding sites when extended under load. The density of integrins can change if a cell is placed on a stiff surface, allowing it to spread, while cadherin density fluctuates significantly during cytokinesis. I considered how changes in the density of stretched integrins or cadherins could pattern an initially random distribution of adhesive molecules. I used statistical field theory methods to find the Ginzburg-Landau free energy of a distribution of adhesive molecules modelled as a lattice gas, and demonstrated that the distributions could undergo two separate phase transitions at low or high density. By investigating the growth of instabilities at the phase transition, I was able to predict the density of clusters as well as the number of molecules per cluster, and found a good match with experimental results. Because clusters of adhesive molecules are essential for cell structure and function, this work was a first quantitative explanation for the initial development of mechanically-induced patterns at the sub-cellular level. **Titin kinase controls muscle growth under load.** Buoyed by this initial success, I shifted my focus during the Covid-19 lockdown to focus in a second stage on the maintenance and adaptation of cell structure to load. Specifically, I was taken aback by the lack of modelling for muscle hypertrophy or atrophy, despite the clear medical and commercial interest in the area. I found it striking that the vast majority of the literature to date focussed on the metabolic contributions to muscle size, even though muscle constituents can locally signal that they are under load. Using the existing literature, as well as structural arguments (e.g. both concentric and eccentric muscle contractions cause hypertrophy), I explained how the part of the titin molecule known as the titin kinase domain (TK) was perfectly placed for muscle mechanosensing. By bringing together the force-length kinetics of TK, the reaction kinetics of the TK signalling pathway, and energy metabolism interactions, I successfully modelled how lengthening TK under load could lead to increased signalling and synthesis of new muscle proteins. I found that mechanosensitive signalling would sharply increase around 70% of the maximum one-repetition force, and that adaptations to non-adaptive resistance exercise plateau after a few months - both of which are observed in the literature. My pioneering work strongly suggests that mechanosensitive signalling controls muscle size, and a medical tool could be developed from this work. This will require training data to be gathered for individuals with different physiological parameters. These ideas can also be applied to the modelling of population dynamics in evolutionary biology which take into account predator-prey interactions that are affected by muscle mechanics, and I look forward to working on this in the future. **Adhesive strength, adaptation, and breaking of myotendinous junctions.** Finally, in a last section, I brought both questions of mechanically-induced structure formation and structure maintenance in biology together. Integrin-mediated adhesion and muscle adaptation to load intersect at the interface between muscle and tendon, the myotendinous junction (MTJ). This region showcases many of the strategies that vertebrates deploy to adapt to load during and after development. Collagen fibres in tendons are very strong but, although they can widen up to an extent, their number does not change after birth. On the other hand, muscle fibres expand very significantly and can fluctuate in size during life depending on exercise and diet. This means that when the force in the muscle increases in tandem with the muscle cross-sectional area, the force per unit area through the tendon and MTJ drastically increases. Although the integrin bonds that hold MTJs together strengthen under load, they do fail past a critical value. Only so many integrins can be packed into an area of membrane, so there is a maximum adhesive force per unit area for a muscle cell. Therefore, the only strategy for the organism to avoid injury is to increase the area of the MTJ. Macroscopic changes to the contact area between muscle and tendon have been reported after exercise, but MTJs are fundamentally constrained by the size of the muscle: e.g., the length and width of the contact area cannot exceed the length and width of the muscle. I used measurements of MTJ area to compare the adhesive force estimated macroscopically from real MTJs with the highest forces experienced in the hamstrings during sprinting. These comparisons showed that many muscles must have a much higher surface area than can be estimated macroscopically. MTJs have solved this surface contact area limitation by developing micrometer scale interdigitations of muscle and collagen at the MTJ. I modelled the position of the muscle-tendon interface in elements of muscle that are connected to collagen fibres and elements that are not, and found that the difference in position can account for the size of the protrusions. These can then lengthen progressively under load as the muscle force and tendon stiffness change during development. This mechanism can explain how these finger-like protrusions are seen to lengthen in exercised rats (and likely in humans too). Fascinatingly, this model predicts limitations of the MTJ area that explain why large animals could be limited in their locomotion (e.g. elephants cannot jump) because of injury potential at the MTJ. I am excited to publish this work, because it shows that integrin and molecules involved in its force chain are a clear target for the prevention of the most common muscle-tendon injuries. **Summary.** My academic style has been to concentrate on writing comprehensive articles on the subjects I have studied in the hope that these will become seminal references in the field, while also writing some shorter papers to present timely side results. I expect my thesis work to motivate better and more numerous experimental studies of adhesive molecular structures, muscle hypertrophy and atrophy, and injuries related to myotendinous junction structure. I hope that it will invigorate a discussion of mechanosensitive structure formation and structure maintenance in medicine and biology, and allow me to capitalise on timely applications of my pioneering models to other areas of biology.
Controlled Access
Phase Transitions in driven 1D and Quasi-Periodic 2D Optical Lattices
Bhave, Shaurya
Ultracold atoms in optical lattices are a versatile experimental platform. They facilitate precisely controlled quantum simulations of a broad range of condensed matter phenomena. This thesis will describe work undertaken with this platform, to explore phase transitions in two different systems. The first is the observation of a first-order Mott transition in a resonantly shaken lattice. The second, is the realisation of a two dimensional Bose glass state in an optical quasicrystal. Quasicrystalline patterns posses long-range order but lack periodicity. This places them in the spectrum of materials between clean, ordered crystals and disordered, amorphous solids. They share properties of both disordered structures and ordered materials, resulting in a rich variety of physical phenomena. This thesis focuses on a novel ground state phase exhibited by our eightfold symmetric optical quasicrystal: the Bose glass. This is a localised phase resulting from the interplay of disorder, in the form of aperiodicity, and interactions. In this regard, I will detail two experiments. The first experiment measures the ground state phase diagram of the quasicrystal as a function of both the lattice depth and interaction strength. The 2D optical quasicrystal is found to support both a superfluid (extended) phase and a Bose glass (localised) phase. The Bose glass phase occurs for lattice depths above a critical point, which gets shifted to larger depths as interactions are increased. This measurement constitutes the first realisation of a 2D Bose glass phase. The second experiment explores the quench dynamics of this phase transition. Rapidly crossing the BG/SF phase boundary induces transient out-of-equilibrium dynamics that depend on properties of the underlying Hamiltonian. This establishes quantum quenches as means to uncover the relaxation dynamics in the long and short term following a rapid change in parameter. The other investigations involve Floquet physics; the study of time periodic Hamiltonians. Introducing periodic time dependence circumvents many challenges of performing quantum simulations with ultracold atoms. Periodic driving introduces a different set of parameters, allowing one to simulate many phenomena that would otherwise be inaccessible. This thesis will present our use of sinusoidal shaking to perform band engineering in a quasi-1D lattice. I will detail our experimental observation of a discontinuous version of the Mott insulator to superfluid transition. This is the first quantum simulation of a discontinuous quantum phase transition in a strongly correlated system. These investigations will open the door to further studies in a 2D square lattice, and eventually the full quasicrystal.
Controlled Access
Constraining Stellar Evolution in Young Open Clusters with NGTS
Smith, Gareth
Understanding stellar evolution is fundamental to astronomy, yet there remain significant gaps in our knowledge. Scientific progress requires testing our best explanations against observation, and open clusters act as key test sites for theories of stellar evolution. Observations of young open clusters enable theoretical models to be assessed at stages when stellar properties still reflect their initial conditions, and in regions of parameter space where constraints remain scarce. The Next Generation Transit Survey (NGTS) is a multi-telescope, ground-based, wide-field photometric survey which is well-suited to studying these environments, and its observations are at the heart of this thesis, which presents my research in the field of early-stage stellar evolution as conducted during my PhD. As a member of the NGTS consortium and, specifically, of the team focusing on open clusters and star forming regions, I have used NGTS data to analyse stellar rotation in the Orion Star-forming Complex and to characterise a low-mass eclipsing binary (EB) in a triple-star system located in the Blanco 1 open cluster. By combining NGTS observations with additional photometry and spectroscopy, and by developing a novel method for extracting radial velocities of close binary (or higher-order) systems, I derived the fundamental parameters of the Blanco 1 EB. With masses and radii measured to a precision better than 1 and 2 per cent, respectively, the newly-identified EB, NGTS J0002-29, becomes a benchmark addition to the current list of 19 well-characterised, low-mass, sub-Gyr, stellar-mass EBs, which constitute some of the strongest observational tests of stellar evolution theory at low masses and young ages. Long-baseline observations of 30 square degrees of the Orion Complex provided simultaneous photometry of thousands of stars, leading to my study of stellar rotation. I analysed cluster membership using astrometry from Gaia DR3 and built a pipeline to process light curves, producing an extensive and homogeneous dataset of more than 2000 rotation periods. I estimated interstellar extinction on a star-by-star basis by fitting broadband photometric data, and derived stellar ages using evolutionary models. I assigned the target stars to kinematic clusters, calculated their ages, and analysed rotation period distributions, finding evidence for mass-dependent evolution during the first 10 Myr, as well as corroboration for the idea that circumstellar discs play a role in regulating the evolution of angular momentum. These studies demonstrate the ability of modern ground-based photometric surveys to address questions about young-age stellar evolution, both on the scale of individual systems and on the scale of large clusters. This ability to operate at different scales facilitates rapid progress and helps to maximise the potential in our data for testing our theories.
Embargo
Spectroscopic Studies of Excited States in Carbene-Metal-Amide Emitters
Reponen, Antti-Pekka
Organic semiconductors are finding increased applications in modern technologies over their inorganic counterparts and hold several advantages from flexibility to better processability. In particular, organic light-emitting diodes (OLEDs) are already available commercially and familiar in consumer applications. Unfortunately, lack of stable and efficient blue-emitting materials has been a long-standing problem which limits the general usefulness of current OLED technologies. As such, development of new materials for and approaches to light emission in OLEDs is a very active area of research. Carbene-Metal-Amides (CMAs) are a recent class of organometallic semiconductors with proposed applications in high-performance OLEDs. Fast triplet harvesting in CMAs occurs through thermally activated delayed fluorescence (TADF). In contrast to usual TADF emitters, CMAs incorporate a metal centre which enhances spin-orbit coupling effects. However, the in-depth emission mechanism in CMAs has been subject to some uncertainty, and high-performing blue CMAs have remained elusive. In this thesis, we use steady-state and time-resolved spectroscopic methods to study excited-state properties in a range of CMA materials. First, using single-atom donor substitutions, we show that energies of multiple states can be simultaneously shifted. We find that these shifts can be qualitatively rationalized from a perturbation theory approach, resulting in a design principle for blueshifting emission while avoiding donor-localized states previously linked to slow emission in blue CMAs. Next, we perform extensive transient absorption characterization of a coinage metal CMA series, using environmental effects to adjust excited-state energies. We find the first direct linkage of excited-state absorption features to specific state characters for CMAs. Results establish groundwork for investigating the population and role of non-emitting ligand-centered states, which have so far remained a source of uncertainty. Finally, we investigate unusual donor fluorescence in CMAs. We evaluate possible sources, such as free donor units and orthogonal conformers. We find several pathways for apparent free donor formation, from degradation to dilution. These results inform us about material stability considerations and advice caution in solution studies of copper CMAs especially.
Embargo
Light and Charge Management in Optoelectronic Systems
Baikie, Tomi; Baikie, Tomi [0000-0002-0845-167X]
This thesis explores the manner in which light interacts with matter, and the subsequent control of photogenerated charge. Firstly, we consider the nature of photogenerated charge in the context of mixed ionic electronic conductors. We introduce a new interpretation of surface photovoltage measurements for these materials and demonstrate control over transport of charge throughout the material. Thereafter, we explore the physical implications of concentrating light. We present a theoretical study detailing it may be possible to concentrate light without the loss of energy by the generation of entropy through photon multiplication. In the context of silicon photovoltaics, we analyse the achievable benefits of light concentration in real world environments. Experimentally, we outline how the effectiveness of concentrating devices may be measured using spatially resolved photoluminescence measurements. Finally, we conduct an extensive ultrafast spectroscopic study exploring the nature of light and charge in a cell. The mathematical basis and workflow is initially introduced, before outlining how ultrafast measurements of live cells can rich information on photoexcited dynamics of living systems.
Open Access
A unified multi-physics formulation for combustion modelling
Nikodemou, Maria; Nikodemou, Maria [0000-0003-4146-5640]
The motivation of this work is to produce an integrated mathematical formulation for the numerical modelling of material response due to detonation wave loading. In particular, we are interested to capture miscible and immiscible behaviour within condensed-phase explosives arising from the co-existence of a reactive carrier mixture of miscible materials, and several material interfaces due to the presence of immiscible impurities such as particles or cavities. The dynamic and thermodynamic evolution of the explosive is communicated to one or more inert confiners through their shared interfaces, which may undergo severe topological change. We also wish to consider elastic and plastic structural response of the confiners, rather than make a hydrodynamic assumption for their behaviour. Recently developed methodologies meet these requirements by means of the simultaneous solution of appropriate systems of equations for the behaviour of the condensed-phase explosive and the elastoplastic response of the confiners. In the present work, we employ a mathematical model proposed by Peshkov and Romenski, which unifies fluid and solid mechanics by means of generalising the concept of distortion tensors beyond solids. We amalgamate this model with a single system of partial differential equations (PDEs) which meets the requirement of co- existing miscible and immiscible explosive mixtures. We present the mathematical derivation of our unified model and construct appropriate algorithms for its solution. The model is extensively verified and validated against exact numerical or analytical solutions and available experimental results. To enable the application of the model to a wider range of problems, we consider two further extensions. The first addresses the extension to multiple inert or reactive materials and its assessment on realistic scenarios like the mechanical sensitisation of liquid explosives. The second is concerned with the use of hyperbolic thermal impulse equations to model the conduction of heat across material interfaces. This is assessed in the scenario of thermal-induced ignition and transition to detonation in explosives. Results of this work indicate that the developed formulation provides a powerful alternative to existing combustion models, able to seamlessly account for the complex and highly non-linear multi-material interactions present in combustion applications.
Open Access
Investigating Ageing of Neurodegenerative Disease Relevant Protein Condensates
Zhou, Alexandra
The phenomenon of liquid-liquid separation (LLPS) has been a popular topic within soft matter physics. There is now a substantial body of literature on biological macromolecules such as proteins that can undergo the LLPS process and form biomolecular condensates. It has been demonstrated repeatedly that these condensates can play a wide range of cellular roles, from transcription to metabolism. In addition, an increasing amount of research is devoted to investigating aberrant protein condensates. The ongoing effort to find the link between dysfunctional proteins condensates and neurodegenerative disease pathogenesis complements the current research. Protein condensates often become pathological when their physical properties change, from a liquid state to a more gel-like or solid-like state. This process is referred to as a solution-to-gel (sol-gel) transition. There are many external factors leading to this property change, such as temperature and stress. There has been as yet no systematic examination of the physical property changes as a function of time, in other words, the ageing of biomolecular condensates. Therefore, this thesis focuses on the ageing behaviour of several neurodegenerative disease-related proteins with a biophysical approach. More specifically, condensate properties including viscosity, complex modulus, and fusion behaviour are investigated and revealed with innovative tools and techniques. At present, little is known about the absolute viscosity of biomolecular condensates both *in vivo* and *in vitro*, especially as a function of time. An initial study using micro-rheology was conducted to acquire viscosity measurements of fresh and aged protein condensates. Moreover, an analysis of the complex modulus of the condensates supplemented our understanding of condensate properties. A combined platform integrating micro-rheology and fluorescence lifetime imaging (FLIM) was then established to take the findings one step further. One difficulty in studying biomolecular condensate properties is the lack of accessibility to condensates in *in vivo* systems. The combined platform was developed to measure both viscosity and aggregation state in *in vitro* environments and provide a relationship between the two quantities. The relationship can then be applied to infer the viscosity or aggregation state of *in vivo* condensates in a situation where only one of the measurements can be obtained easily. Such a platform provides better insight into the ageing of condensates in a biologically relevant context. To further improve the understanding of biomolecular condensates, this thesis also encompasses an investigation of the physical interaction between condensates. Leveraging a dual-trap optical tweezer setup with bright-field and fluorescence imaging and force response probes, the fusion behaviour of protein condensates was studied in detail. Bright-field microscope allowed protein fusion driven by surface tension to be observed and recorded, and force measurements reinforced the findings with highly accurate readouts on the characteristic fusion time. Two-channel confocal imaging highlighted new and interesting observations on how fresh and aged condensates could influence each other. This thesis may contribute towards a better understanding of biomolecular condensate ageing, especially the ones that are accounted for prominent neurodegenerative diseases when becoming aberrant. Building upon the tools and techniques developed within this thesis, further studies on a broader range of biomolecular condensates can reveal additional implications of condensate ageing in biology and medicine.
Embargo
Machine Learning for Structural Characterization and Generation: Applications to Small-Angle Scattering and Electron Microscopy
Yildirim, Batuhan
The era of big-data-driven science has brought to light the need for new methodologies to process and extract otherwise-unattainable insights from the vast amounts of data generated by materials and nanostructural characterization methods. Small-angle scattering (SAS) and electron microscopy (EM) experiments yield rich data sets that contain structural and morphological information of nanostructures. Exploiting data-driven methods to extract these insights is a natural fit. Additionally, the volume of data produced in the previous computational and simulation age of science has led to the establishment of extensive repositories of structures. These resources present opportunities for functional-property prediction to reveal novel uses for existing structures and for deep generative models to design new structures for a wide range of applications. This thesis is concerned with the development of machine learning (ML) algorithms for the characterization and generation of materials and nanostructures. Chapter 1 discusses characterization and generation in the data-driven age of science and reviews the application of ML to aid these processes, with a focus on SAS and EM for characterization. Chapter 2 provides a technical outline of ML in general, including the specific methods that are employed in subsequent chapters to develop models for processing SAS and EM data for characterization, as well as atomic structures for property prediction and generation. In Chapter 3, a convolutional neural network-based segmentation algorithm is developed to detect and locate nanoparticles in EM images. This constitutes the particle segmentation module of ImageDataExtractor, an open-source software tool developed therein for extracting information from EM images in an automated fashion. Techniques from Chapter 3 are employed in Chapter 4 to calculate SAS intensity functions from morphological and position information of nanostructures obtained from 2-dimensional EM images. In Chapter 5, the focus shifts to SAS, where a multi-task neural network is developed to concurrently identify the size and shape-parameters of the scatterers that produced a given SAS intensity. Chapter 6 constitutes the property prediction and structural generation portion of this thesis, in which a deep generative model of inorganic crystal structures is developed alongside a graph neural network to predict the properties of the generated structures. Finally, concluding remarks and avenues for future research are discussed in Chapter 7.
Embargo
Spin dynamics of bulk and quasi-two-dimensional metal halide based perovskite semiconductors
Bourelle, Sean
This dissertation presents the ultrafast analysis of the electronic states within solution processed bulk and quasi-two-dimensional (quasi-2d) metal halide perovskites. Specifically, of their spin dependencies on lattice and many-body interactions. Many of these results were obtained in close collaboration with others, as specified throughout the dissertation. First, we report that spin-dependent exciton interactions modify the exciton energy and induce spin depolarisation in the quasi-2d Ruddlesden-Popper hybrid metal-halide perovskite Butylammonium Formamidinium Lead Iodide, BA₂FAPBI₇. Experiments were carried out jointly between Dr Ravichandran Shivanna and me. For carrier densities above ~10¹⁵ cm^-3, the room temperature exciton depolarisation rate increases with exciton density, showing that many-body interactions outcompete the precessional motional narrowing spin relaxation that was previously reported for lower carrier densities. Therefore, we observe that room temperature spin lifetimes are largest, ~3.2 ps, at an operating carrier density of ~10¹⁵ cm^-3. We also demonstrate the dynamic circular dichroism that arises from a photoinduced polarisation in the exciton distribution between total angular momentum states. We rationalise this dichroism by a polarisation-dependent exciton-exciton interaction energy that splits co and counter polarised states by 25 meV. In the second part of the thesis, we analyse spin relaxation at cryogenic temperatures, and investigate how the coupling between excitons and the soft ionic crystal lattice impacts the spin relaxation of exciton states in BA₂FAPbI₇. Faraday rotation measurements were obtained by Dr Soumen Ghosh and Dr Franco V. A. Camargo in close collaboration with me. We report that photoexcitation with energy in excess of the exciton absorption peak increases the spin lifetime by two orders of magnitude at 77K. Our hypothesis is that such optical control of the spin relaxation mechanism arises from the strong coupling between excitons and the optically excited phonons, which form polarons at low temperatures that experience a different spin depolarisation mechanism. This new spin-control phenomenon highlights the special role of exciton-lattice interactions on the spin physics in the layered perovskites, and the potential to use different wavelengths of light to encode spin information. In the third section of this thesis, we explore the potential of a two-photon transient absorption technique to analyse the change in optoelectronic properties as a function of film depth and measurably determine the impact of the direct-indirect nature of electronic states in hybrid metal-halide perovskites on charge-carrier dynamics. Experimental work and subsequent analysis in this section was completed jointly between Dr Thomas Winkler and me. We selectively excite surface and bulk states of bulk MAPbBr₃ and employ one and two-photon transient absorption spectroscopy to reveal the energetic landscape. We show that the surface region is greater in band gap energy and shows reduced Rashba-type band splitting. Carriers diffuse away from trap-rich surface states and recombine efficiently within the bulk, despite the significant band splitting that has been argued to inhibit efficient radiative recombination. First first-principles calculations performed by Dr Xie Zhang resolve this apparent controversy and are included within the thesis to demonstrate how bright emission arises from the bulk.
Open Access
Quenching normal Bose gases to Unitarity
Man, Jay
The presence of inter-particle interactions in a system elevates the physics involved from fundamentally single-body, with each particle following its own trajectory according to external forces, to many-body, with collective behaviour emerging from the interplay between the multitude of particles. Furthermore, the stronger the interactions, the more richly many-body the behaviour becomes; weak interactions can be modelled as a background ‘mean field’ whereas the presence of strong interactions introduces fluctuations and correlations that cannot be simplified in this fashion. This Thesis is concerned with two experiments utilising tunable interactions in three-dimensional thermal Bose gases, with particular emphasis on the unitary regime of maximal interactions. We use 39K, a bosonic isotope, in anisotropic harmonic optical trapping potentials. In the unitary Bose gas we find many-body complexity introduced by the possibility of three particles coming into close proximity. However, the preparation of Bose gases with such strong interactions is hampered by this, due to destructive three-body recombination events which eject particles and heat the cloud. To mitigate this, we perform a spin-flip that changes the internal atomic state from one with weak intra-state interactions to one with strong intra-state interactions. This technique constitutes an interaction quench since it takes place on a short timescale compared to other evolution timescales of the ultracold cloud. One experiment concerns the hydrodynamic expansion of a gas after release from an anisotropic trapping potential. When interactions are sufficiently strong, a pronounced inversion of the anisotropy during expansion can be observed which is a manifestation of interaction-driven collective flow. We show that this elliptic flow is intimately linked to thermalisation, and show that it is highly dependent on the microscopic details of the collisions involved. The other experiment addresses the two- and three-body contacts in the Bose gas. These contact parameters arise from the corresponding two- and three-body correlations present in Bose gases, and underpin a large number of macroscopic thermodynamic variables. We map out the two-body contact across the full range of interactions from weak to strong, including the unitary regime, and reveal the three-body contact at unitarity.
Embargo
Charge Transport in Field-Effect and Ion-Gel Gated Conjugated Polymers
Zhang, Lu
Organic field effect transistors (OFETs) have recently made remarkable advancements, which have mostly been driven by the design, characterisation, and comprehension of new conjugated polymers. Significant effects can arise from making minor alterations to the monomers and building blocks, which impacts charge transport; however, some of mechanisms responsible for the effects have yet to be fully understood. There have been simultaneous interests in devising various arrangements of co-monomeric donor−acceptor (DA) copolymers. To create the next generation of high-performance organic semiconductors, it is essential that the relationship between structures and properties of conjugated systems are fully understood. To investigate diverse, customised conjugated polymers with a low degree of disorder, 15 derivatives of the benchmark low disorder polymer semiconductor IDT-BT were synthesised. To create the derivatives, a number of modifications were made to the acceptor unit, donor unit, molecular weight or side chains. The modifications resulted in two polymers, IDT-BS and TIF-BT, which when properly optimised, have the potential to perform better than IDT-BT. Thus, these novel polymers are suitable for use in flexible electronic devices. Moreover, the properties of semiconducting organic materials can be altered by chalcogen substitution. In this thesis, we not only explore the effects of chalcogens on the field-effect charge transport properties of such materials but also examine the effect of chalcogens upon doping. It is considered that interchain interactions are enhanced by introducing a single chalcogen atom into the backbone of the polymer. As a result, the doped samples adopt better electrical properties. In our opinion, the enhanced properties are the product of the heteroatom stimulating the formation of unique microstructures. We also investigate the impact on conductivity caused by three different doping methods (molecular doping/ ion exchange doping and electrochemical doping). We discovered that electrochemical doping based on organic electrochemical transistor devices is one of the most effective ways to continually modify the conductivities by applied voltage with incredibly high doping levels (one charge per monomer). The potential for future double-gating investigations is provided by this reversible and stable procedure. Furthermore, this thesis not only discusses the performance but also the charge transport mechanism of conjugated polymers. We provide a novel experimental method for directly probing the density of states close to the Fermi level without changing the shape of the density of states (DOS) based on dual-gated organic electrochemical transistors with an electrolyte top-gate and a dielectric bottom gate. This new technique suggests a strategy for modifying carrier concentration while simultaneously employing electrochemical and field-effect gating to calculate mobility in semiconducting polymers. This is due to the fact that, in contrast to electrochemical gating, field-effect gating is manageable and sufficient to enable the modulation of a defined population of carriers, leading to an accurate measurement of field-effect mobility. According to our knowledge, this double-gating research is the first to use this technique to investigate field-effect charge transport at various strongly doped states and link field-effect transport to the Seebeck coefficient's sign changes. Last but not least, we studied five low band-gap open-shell co-polymers. This project is evident that high temperature annealing has an impact on both mobility and microstructure, proving that interchain ordering is crucial for charge transport in open-shell polymers. Therefore, it can be assumed that increasing charge transfer at both the intrachain and interchain levels will improve both the hole and electron mobilities of open-shell polymers. As a result of our research, we think it may be possible to create more stable open-shell polymers with higher mobility that may provide new possibilities for investigating the physical phenomena associated with spin in polymer semiconductors.
Embargo
Mixed Lead-Tin Halide Perovskites for Optoelectronic Applications
Dey, Krishanu; Dey, Krishanu [0000-0003-3469-6184]
Mixed lead-tin (Pb-Sn) perovskites are unique materials in the family of halide perovskites. Unlike Pb perovskites, these mixed-metal systems can demonstrate bandgaps below 1.3 eV and are therefore essential constituents for low bandgap bottom subcell in all-perovskite tandem solar cells as well as for near-infrared light emitting diodes (LEDs), lasers and photodetectors. Although the air stability of these Sn-containing perovskites are relatively poor due to the facile oxidation of Sn2+ to Sn4+, these materials do possess certain bright aspects in their optoelectronic properties which have received less attention in the community and this forms the foundation of this thesis. Chapter 1 provides a bigger picture of the need to explore sustainable alternatives to energy generation and consumption and the role of emerging semiconductor materials, especially metal halide perovskites, in that pursuit. Chapter 2 provides a general background to semiconductors and outlines the operating principles of solar cells and FETs. It also presents the current understanding of the optoelectronic properties and degradation mechanisms of mixed Pb-Sn halide perovskites. All the experimental techniques used in the thesis are introduced in Chapter 3. Chapter 4 summarises the optimization strategies of mixed Pb-Sn halide perovskite systems for demonstrating reliable and hysteresis-free p-type perovskite FETs with high hole mobility reaching 5.4 cm2/Vs and ON/OFF ratio approaching 106, which are among the best metrics in the field of perovskite FETs. We also rationalize these findings of long-range lateral transport with the support of theoretical calculations, film morphology studies and chemical analysis of defects in these materials. We then extend the above work to probe the lateral charge transport mechanism in mixed Pb-Sn perovskite FETs in Chapter 5. Through temperature-dependent field-effect mobility measurements, aided further with photoluminescence microscopy under bias, we show that ionic screening effects are greatly suppressed in mixed Pb-Sn devices when compared to their Pb-based analogues. We also demonstrate that dipolar disorder (associated with methylammonium, MA+ cation) induced lowering of FET mobility near room temperature can also be seen for mixed Pb-Sn perovskites and hence further efforts need to be invested in going MA-free in future. Next, we generalize the above findings of suppressed ion dynamics in mixed Pb-Sn systems by fabricating optoelectronic device stacks with vertical charge transport in Chapter 6, which are relevant for solar cells and LEDs. We reconcile these findings through first principles calculations, which reveal the key role played by Sn vacancies (with low formation energy) in increasing the migration barrier for iodides due to severe local structural distortion in the lattice. In Chapter 7, we show that the partial or complete incorporation of Sn in the metal (B) site of mixed halide perovskites offer very promising intrinsic stability to halide segregation under a host of processing and operational conditions. We further study the optoelectronic properties of these mixed halide Pb-Sn perovskites to understand the impact of light soaking on the charge carrier recombination and transport in these materials. We also assess the device performance of these mixed halide perovskite materials by fabricating single single junction solar cells. Chapter 8 summarises the key findings of this thesis and proposes several potential directions of research involving these mixed Pb-Sn perovskites. All the work presented herein provides an important advance to the fundamental understanding and applied device integration of mixed lead-tin perovskite materials and can be leveraged for demonstrating a ‘perovskite optoelectronic universe’ with high performance and stability.
Embargo
Helium Atom Scattering from Chiral Structures
Eratam, Fulden
Helium Atom Scattering (HAS) is the only diﬀraction technique that combines absolute sur- face sensitivity with non-destructiveness and universality, but has been seldom applied to the study of chiral surfaces. The current thesis focuses on advancing the available theoretical and experimental tools to motivate the study of chiral surfaces using atomic beam techniques. The importance of chirality is discussed in chapter 1, along with the signiﬁcant role played by surfaces in the creation of chiral media required for biologically and industrially relevant enantioselective reactions. The key aspects of the HAS technique is introduced and a candidate system, namely D-alaninol adsorbed on Cu(100), is proposed for the exploration of chiral expression on a metal surface. The research involves a two-fold approach: First is instrumental considerations to im- prove data acquisition and analysis; the second is the development of a theoretical basis to help quantify the interaction of He with a chiral surface as well as assess its sensitivity as a probe of surface chirality. Chapter 2 introduces a 3D scattering simulation that models the scattering apparatus, to com- pute diﬀraction peak proﬁles likely to arise from a periodically arranged plane of point scatterers representing the sample. A structural analysis of the multi-layer adsorption of D-alaninol on the Cu(100) surface has been provided in chapter 3. The experimental data was acquired using the MiniScat spectrometer. In addition to the 1D and 2D diﬀraction spectra arising from the chiral system, the uptake and desorption behaviour of the chiral D-alaninol molecules have been inves- tigated. In general, the experimental data was found to be in good agreement with the published data acquired with other surface techniques such as LEED, STM and XPS. To overcome the relative complexity of the chiral organic/ metal interface, the He-D-alaninol interaction was ﬁrst modelled on another methylated system that was simpler. Therefore, an in- teraction potential function originally applied to the diﬀerential cross-section analysis of crossed atomic and molecular beams has been proposed and tested on the CH3-Si(111) surface, as de- scribed in chapter 4. Through the close-coupled analysis performed on the system, it became possible to assess the level of transferability between an interaction model describing the scat- tering of thermal He atoms by a crossed atomic-molecular beam and a second model describing He scattering by a gas-phase adsorbate. The close-coupled analysis was repeated for the D-alaninol/Cu(100) system in chapter 5, using both an asymmetrically corrugated and a pairwise version of the interaction potential previ- ously employed. The level of agreement between the experimental diﬀraction spectra and the close-couple computed diﬀraction data was assessed and a superior hybrid potential model was introduced. Considering the two-element adsorbates typically studied, the relatively large ad- sorbate unit cell size of the D-alaninol/Cu(100) system makes it one the most complex organic systems where the close-coupled approach has been successfully applied. A new ion-source design has been proposed and characterised in chapter 6. The upgrade resulted in 3 orders of magnitude increase in the detector sensitivity relative to the commercial quadrupole analyser previously installed. The ion-source and the subsequent ion-optics elements of the new detector assembly has been modelled using an existing Boris algorithm, as described in chapter 7. Based on the simulation data, practical improvements oﬀering another order of magnitude increase in the detector eﬃciency has been identiﬁed. Finally, in chapter 8, a future direction for research on chiral surfaces using atomic beam techniques has been proposed.
Open Access
Dynamics of frustrated magnetic systems – Emergent fractals and anomalous magnetic noise in spin ice
Nilsson Hallén, Erik Jonathan; Nilsson Hallén, Erik Jonathan [0000-0003-4883-4832]
Typical thermodynamic systems have a small number of minimum-energy ground states into which the system strives to arrange itself when the temperature is sufficiently low. In some cases, however, competition between interactions leads to the suppression of conventional long-ranged magnetic ordering, through a mechanism known as frustration. One of the typical features is the appearance of a large number of degenerate (or nearly degenerate) minimal energy states, that can exhibit topological order. The dynamical properties of topological systems, in particular those that host fractionalised excitations, is currently a topic of intense research activity. This thesis focuses on one specific instance of a topological frustrated magnetic system: spin ice. Spin ice is one of the most thoroughly researched frustrated models, exhibiting exciting phenomena typical of topological matter – including an emergent gauge field description with fractionalised excitations that take the form of emergent magnetic monopoles. Highly accurate numerical methods and analytical techniques have been developed to model spin ice materials, and have facilitated a detailed understanding of their static properties. The primary aim of this thesis is to address some of the outstanding questions about the dynamics of classical spin ice materials. The focus will be on two dynamical properties: the magnetic relaxation time and the magnetic noise spectrum. Several independent experiments, some dating back over two decades, have found that the magnetic relaxation time in these materials diverges rapidly upon cooling – more rapidly than what any previous theory can explain without invoking extrinsic contributions such as surface effects, disorder, and temperature dependent microscopic time scales. Previous theories also predict that the magnetic noise spectrum of spin ice should be Lorentzian, with a power spectral density that decays as ν^(-2) with the frequency, ν. Here, noise spectra measured on a single crystal of Dy₂Ti₂O₇ are presented and shown to instead decay as ν^(−α), with an anomalous exponent α ≈ 1.5. Through a combination of extensive numerical modelling and fundamental arguments about the motion of the magnetic monopoles, I establish that the conventional model of spin ice dynamics cannot explain the anomalous magnetic noise. I resolve these issues by introducing a new model for the dynamics, which includes the effects of local symmetry on the flip rate of spins. The main insight the new model leads to is the emergence of dynamical fractal structures on which the magnetic monopoles are constrained to move. It is by hosting the monopole motion that the fractals bequeath the magnetic noise with an anomalous exponent and cause a slow-down of the relaxation. The spatial properties of the fractals thus manifest themselves in the time dependence of the magnetisation – and the anomalous dynamics observed in Dy₂Ti₂O₇ are evidence of fractal objects in a disorder-free, bulk crystal. Indeed, the new model dynamics quantitatively captures the experimentally observed noise spectrum and relaxation times, without the introduction of any additional fitting parameters. The dynamical fractals cannot be imaged directly, but do influence other dynamical properties, besides the already discussed magnetic noise and relaxation time. Using simulations and simplified models of analogous systems, I investigate how the fractals affect both equilibrium and out-of- equilibrium transport properties. The response of spin ice to an oscillating magnetic field is identified as one particularly promising area to explore in future experiments, with suggestions made for the field and temperature regimes in which fractals are likely to manifest themselves. An alternative spin ice model capable of reproducing the experimentally observed relaxation time and anomalous magnetic noise, but incompatible with the thermodynamics of spin ice materials, is also introduced. The model includes a specific choice of third-nearest-neighbour exchange interactions, which are frustrated independently of the usual spin ice interactions. These additional interactions are found to induce a phase transition between the normal spin ice phase and a spin nematic phase – a phase displaying rotational symmetry breaking without long-ranged order. In the “nematic spin ice” phase, low energy paths are formed along which magnetic monopoles move freely. These paths form anisotropic fractal clusters with properties akin to the dynamical fractals mentioned previously. Thus, the anomalous noise spectrum is again found to originate from the motion of magnetic monopoles on fractals.
Open Access
Harnessing Vibrations for Efficient Exciton Dynamics in Semiconducting Energy Materials
Sneyd, Alexander; Sneyd, Alexander [0000-0002-4205-0554]
This dissertation describes our study of the fundamental role vibrations play in the excited-state dynamics of semiconducting energy materials. We examine these effects in self-assembled organic semiconducting nanostructures and small molecules, focussing on the implications for exciton transport, energy transfer, and light emission. Special use of ultrafast laser spectroscopy techniques such as impulsive vibrational spectroscopy and transient absorption microscopy is made to directly observe vibronic couplings and exciton transport. In self-assembled poly(3-hexylthiophene) nanofibers we observe exceptional exciton transport that cannot be explained with current models of exciton transport, despite low energetic and structural disorder. By directly measuring the excited-state vibrations, we are able to construct non-adiabatic simulations which reveal that zero-point motion enables access to delocalized states which mediate transport. This new transient delocalisation mechanism of transport can enable higher efficiencies and new device architectures. We follow this up by combining polyfluorene nanofibers with inorganic quantum rods for the purpose of energy transfer, and observe high levels of energy funnelling to the rods. Such behaviour has strong prospects for multielectron photocatalysis and upconversion. Finally, we assess the role of vibrations in the emission dynamics of several archetypal thermally-activated delayed-fluorescence emitters. We reveal their excited-state vibrations and track changes over time due to environmental relaxation. This serves to rationalize favourable emission bandwidths, low Stokes shifts, low non-radiative rates, and spin-orbit coupling enhancements. Our results challenge current pictures of exciton dynamics, and assert the varied and profound role vibrations have on properties such as energy transport and light emission. Traditionally, the uniquely strong vibrational couplings of organic semiconductors have been thought of as deleterious, but here they present themselves as an asset. For exciton transport especially, we propose design rules to harness vibrations which may enable the next generation of efficient optoelectronic devices.

Browse

Recent Submissions