Search

Environmental and pathogenic Gram-negative bacteria are exposed to lipid peroxidation emerging from oxidative stress conditions. However, the mechanism to survive this stress is essentially unknown. The opportunistic pathogen Burkholderia cenocepacia encodes three genes that are upregulated under oxidative stress conditions such as sub-lethal concentration of antibiotic and superoxide ion stress. One of these genes, bcnA, encodes a periplasmic protein from the lipocalin family, which is usually associated with a gene encoding a cytochrome b561 inner-transmembrane protein that we named lcoA; this synteny is conserved in the majority of the β-proteobacteria. Compared to the wild type, mutants lacking these proteins and psrA, a gene encoding a cytoplasmic aldehyde reductase that reduces toxic aldehydes, are more susceptible to bactericidal antibiotics and produce more lipid peroxidation by-products. Fluorescent microscopy analysis using a probe that interacts with lipid peroxyl radicals shows an accumulation of these molecules at the bacterial cell poles and septum. Additionally, using a dye that interacts with anionic phospholipids, we found that peroxidation is associated with a redistribution of anionic phospholipids in the membrane. From this, we conclude that BcnA, LcoA, and PsrA are components of an evolutionarily conserved, previously obscure peroxidation detoxification system that protects the bacterial cell envelope from lipid peroxyl radicals.

This dissertation investigates the process by which urban planning agents imagine futures and the sociotechnical conditions which help establish the credibility of their expectations. Aiming to shed light on the practice of city-making, the thesis focuses on how urban planning agents co-produce specific urban-regional sociotechnical imaginaries. Based on the assumption that city-regional innovation processes are inherently future-oriented and imply a profound reordering of society, the thesis pays close attention to who gets to construct and diffuse imaginaries of desirable futures. Drawing on documentary research, 50 interviews and ethnographic observations, this study empirically engages with the implementation of a contemporary urban-regional redevelopment project, the Edinburgh and South-East Scotland City Region Deal (CRD). A £1.3bn infrastructure investment by the Scottish and UK governments and local partners over fifteen years, it is designed to accelerate data-driven innovation and inclusive growth. Officially signed in August 2018, the CRD represents a unique case study for at least two reasons: first, it constitutes the first higher education-driven city region deal in the UK. Second, it aligns a cluster of considerably powerful actors - political leaders, higher education officials, tech entrepreneurs - around one specific sociotechnical imaginary: establishing the city region as the "Data Capital of Europe". Conceptually rooted at the intersection of Science and Technology Studies (STS) and Urban Studies, this dissertation argues that future-making in the Edinburgh city region is characterised by an unequal distribution of projective agency and driven by an entrepreneurial class at the nexus of science and politics. Shedding light on the complex interplay of expectations and expertise in urban-regional development, this study demonstrates the growing influence of global higher education institutions such as the University of Edinburgh in shaping desirable futures. More specifically, the concept of "dataist modernity" is developed to characterise a form of modernisation that is rooted in an ontology of knowability and overestimates the power of data in solving complex social problems. By crafting this line of argument, this study extends co-productionist STS on 'smart urbanism' to take fuller account of the intricate relationship between technoscience and urban-regional planning.

Planets are born in protoplanetary discs (PPDs), made of gas and dust, which surround newly formed stars. Gravitational coupling provides planets with means to interact with their birth environment, leading to the excitation of global spiral density waves, which drive evolution of the disc (such as the opening of gaps) and the planets' orbital properties (known as migration). Modern observations of PPDs, provided for example by the high resolution interferometer ALMA, have shown intriguing features in continuum dust emission, such as lobes and arcs neighbouring darker ring-like gaps. These have been interpreted as dust-loaded vortices, which might be efficiently formed via the instability of the edges of planet-driven gaps. The main motivation of this thesis is to improve our understanding of their formation in order to put constraints on yet unseen planets. Consequently, in this thesis, we investigate this interaction of young planets with the disc gas they are embedded in, focussing on the non-linear shock-driven damping of global waves, subsequent disc evolution and the stability of the resulting flow. We perform a detailed numerical study of this process at the highest spatial resolution to date, investigating the formation, propagation and shock-driven damping of global density waves. The high fidelity of our numerical models allows us to thoroughly test previous theories of weakly nonlinear wave propagation due to Goodman and Rafikov (2001) and Rafikov (2002b), finding agreement with simulations on a qualitative but not quantitative level. As we show, this shortcoming is mainly due to the fact that this theory does not include the dispersion of shearing waves in differentially rotating discs. We provide fitting formulae for characteristics of the shocking wave and use it to explain observed vortensity production the closely related opening of a gap. This detailed study allowed us to understand how planet masses and disc surface density slope control this process from first principles. One of the main results of this work is a simple one-dimensional framework, which uses these fits in order to efficiently construct time-dependent gap models for various disc and planet parameters. Our model is verified against long-term numerical simulations and we use it further to perform a linear stability analysis of the gap edges against the Rossby wave instability (RWI). We find a power-law like behaviour of time-scales of instability and subsequent vortex formation, which are both strong functions of planet mass. Again, we provide useful fitting parameters. Further, we present a new method of constraining the age and masses of putative planets in observed gaps, depending on the presence or absence of vortex signatures. Being very sensitive to the disc temperature, such constraints will become more stringent as measurements of this parameter improve. Lastly, a detailed study of the more global structure of density waves and their gravitational coupling to the planet is presented. We investigate a repetitive pattern in the excitation torque density, which has been previously observed in simulations. We show that this phenomenon is indeed physical, giving a detailed explanation of its origin and reason why it only occurs in the outer disc. By varying the disc model, we show that the phenomenon occurs using barotropic and adiabatic equations of state, as well as in discs that allow for thermal relaxation. It is also robust against some wave nonlinearity such that our results may be applied also to other systems, such as circumbinary discs, as we show.

This thesis reflects on how to assess unfair inequality in health. Its main arguments are that the Equality of Opportunity framework may be more adequate for this task compared to the framework of socioeconomic health inequalities, but that there are several limitations regarding how the Equality of Opportunity framework has been implemented. The thesis features a systematic literature review and five empirical applications, including three questionnaire-experimental studies. The literature review in Chapter 2 explores how the framework of Equality of Opportunity has been used to assess health inequalities. The third chapter looks at socioeconomic health inequalities in Chile and it shows for the first time that socioeconomic differences in life expectancy have increased in the last two decades. Four limitations of the theory of Equality of Opportunity are explored. Chapter 4 argues that the 'control approach' has failed to provide an operational definition of effort. In this respect, it was found that the requirements needed to provide a definition of effort according to such an approach are not satisfied empirically. Chapter 5 shows that there are several challenges to compensate for the indirect effect of circumstances. It was found that an alternative compensation criteria has moderate acceptance among Chilean members of the public. Chapter 6 uses the theory of equivalent income to provide a basis to reduce inequalities among individuals who face the same constraints. This idea finds some support among a representative sample of Chilean adults. Chapter 7 suggests that the 'preference approach' may be a better strategy to operationalise effort and circumstances. In this chapter an empirical study assesses Equality of Opportunity according to this approach. The thesis concludes with the proposition that the preference approach and the theory of equivalent income may offer a better alternative to assess unfair inequalities.

Economists and policy makers are increasingly concerned with the heterogeneous impact of economic policies, on top of their overall effectiveness. However, understanding precisely how they impact different population groups often requires specific approaches and data sets. This thesis contributes to this issue by leveraging novel data sets and methods to improve the economic analysis of carbon emissions, the measurement of inflation and the measurement wage inequality. The first chapter analyzes to what extent countries can reduce their CO2 emissions by changing the composition of consumption rather than the underlying technology, with a focus on food products. I document that carbon intensity is heterogeneous even within detailed product categories. I then show that well-targeted taco deliver large efficiency gains, and that the impact of carbon taxes across households varies strongly with their exposure to high-carbon products, but not with their expenditure level. In addition, the welfare cost of reducing carbon emissions varies strongly across product categories, providing a new justification for granular taxes or, equivalently, carbon markets. The second chapter concerns the inflation dynamics during the Covid-19 lockdown in France. Using scanner data on fast-moving consumer goods from a large retailer, I find that the lockdown lead to an important, generalized but temporary increase in price levels across product categories. Further, I find that this inflation shock was asymmetric across products, households and cities, and that this asymmetry did not vanish on the medium-term. The third chapter focuses on the contribution of firm heterogeneity to wage inequality and its measurement in two-way fixed effect models. I provide evidence that firm-side drivers of wage inequality are overestimated by at least 25% because of model overfitting. I then provide a simple procedure to recover the correct measures of interest, show that the correction matters quantitatively and derive more precise estimates of firm effects using shrinkage methods.

Party organisations are no longer limited to the domestic arena. A growing number of works has examined why parties become transnational actors, and how they campaign, and organize abroad. However, still largely unexplored is how parties become and remain transnational actors. Existing work investigates either the transnational activities of party organisations, or migrants, but not the interaction of both. The mechanisms that underpin the transnationalisation of political parties are therefore not fully understood yet. This thesis addresses this gap. I argue that parties must actively negotiate their entry and presence within migrant communities. I claim that emigrants often oppose the presence of parties, but that parties can partly overcome this resistance with a specific set of incentives, mobilization, recruitment, and linkage strategies. I situate my analysis in the relevant transnationalism and political science literatures. To empirically support my argument, I draw on 62 interviews with hard-to-reach key informants and 50 logged field reports, which I conducted during 13 months of ethnographic fieldwork in Barcelona. The comparative research design includes Bolivia, Ecuador, and Peru. First, I find that parties succeed as transnational actors when migrants can use them to produce collective goods in the home and host country. Migrants also engage when parties can stimulate emotional bonds, or offer political career opportunities. Second, successful transnational parties exploit informal networks and harness networks and organizational structures of local migrant organisations and Spanish parties. Finally, migrant organisations act as strong gatekeepers and force party chapters, who lack funding and political clout, to apply more aggressive, informal, and multi-layered linkage tactics. This thesis contributes to our understanding of transnational politics by providing first-hand, in-depth insights into how political parties establish a presence abroad, the challenges that they face, and how both home and host country factors shape this development.

Cooperation between humans has been foundational for the development of civilisation and yet there are many questions about how it emerges from social interactions. As artificial agents begin to play a more significant role in our lives and are introduced into our societies, it is apparent that understanding the mechanisms of cooperation is important also for the design of next-generation multi-agent AI systems. Indeed, this is particularly important in the case of supporting cooperation between self-interested AI agents. In this thesis, we focus on the analysis of the application of mechanisms that are at the basis of human cooperation to the training of reinforcement learning agents. Human behaviour is a product of cultural norms, emotions and intuition amongst other things: we argue it is possible to use similar mechanisms to deal with the complexities of multi-agent cooperation. We outline the problem of cooperation in mixed-motive games, also known as social dilemmas, and we focus on the mechanisms of reputation dynamics and partner selection, two mechanisms that have been strongly linked to indirect reciprocity in Evolutionary Game Theory. A key point that we want to emphasise is the fact we assume no prior knowledge and explicit definition of strategies, which instead are fully learnt by the agents during the games. In our experimental evaluation, we demonstrate the benefits of applying these mechanisms to the training process of the agents, and we compare our findings with results presented in a variety of other disciplines, including Economics and Evolutionary Biology.

Discovering hidden partial differential equations (PDEs) and operators from data is an important topic at the frontier between machine learning and numerical analysis. Theoretical results and deep learning algorithms are introduced to learn Green's functions associated with linear partial differential equations and rigorously justify PDE learning techniques. A theoretically rigorous algorithm is derived to obtain a learning rate, which characterizes the amount of training data needed to approximately learn Green's functions associated with elliptic PDEs. The construction connects the fields of PDE learning and numerical linear algebra by extending the randomized singular value decomposition to non-standard Gaussian vectors and Hilbert--Schmidt operators, and exploiting the low-rank hierarchical structure of Green's functions using hierarchical matrices. Rational neural networks (NNs) are introduced and consist of neural networks with trainable rational activation functions. The highly compositional structure of these networks, combined with rational approximation theory, implies that rational functions have higher approximation power than standard activation functions. In addition, rational NNs may have poles and take arbitrarily large values, which is ideal for approximating functions with singularities such as Green's functions. Finally, theoretical results on Green's functions and rational NNs are combined to design a human-understandable deep learning method for discovering Green's functions from data. This approach complements state-of-the-art PDE learning techniques, as a wide range of physics can be captured from the learned Green's functions such as dominant modes, symmetries, and singularity locations.

Go is a programming language that has gained increased popularity due to its good support for system programming and its channel-based message passing concurrency mechanism. These features rendered Go the language of choice of many platform software developers. Go offers a wide range of primitives to coordinate lightweight threads, e.g., channels, waitgroups, and mutexes. Although, these concurrency primitives help mitigate data races, they introduce additional complications due to the complexity of reasoning about concurrency. In this thesis, we first perform an empirical analysis on concurrent Go programs which analyses 125 Go projects from GitHub in order to understand how concurrency is used in publicly available code. Our results include the following findings: (1) concurrency primitives are used frequently and intensively, (2) most projects use synchronous communication channels over asynchronous ones, and (3) most Go projects use simple concurrent thread topologies, which are however currently not fully supported by existing static verification frameworks. To address these limitations, we propose a novel static checker for Go programs that relies on performing bounded model checking of their concurrent behaviours. In contrast to previous works, our approach deals with large codebases, supports programs that have statically unknown parameters and is extensible to additional concurrency primitives. Our work includes an empirical analysis that studies the usage of concurrency in Go projects, a detailed presentation of the extraction algorithm from Go programs to Promela models, an algorithm to automatically check programs with statically unknown parameters, and a large scale evaluation of our approach. The latter shows that our approach outperforms the state-of-the-art.

Identifying the mechanism by which flow transitions to turbulence is of particular interest for the design and control of atmospheric flight vehicles. Compression corners are present on many key components of aerospace vehicles whose performance is essential for safe operation. Therefore, it is crucial to understand the transition mechanisms to avoid excessive thermal loading of the vehicle which is caused by separated flow and shock boundary layer interaction. The transition mechanisms involved in many high-speed separated boundary layer scenarios, are not fully understood. Linear stability theory, developed in the last century, describes the decomposition of a flow into a basic state and a small amplitude disturbance. The behaviour of these disturbances can be described as damped or growing. When growing disturbances are uncovered, these will begin to grow exponentially, leading to transition. This behaviour is known as modal. When a flow is subject to high external disturbance levels, it may cause a bypass of the modal amplification scenario, leading to the non-modal stability becoming a path to transition. In order to truly represent flows for the purposes of stability analysis, highly accurate base states are required. Several methods for this are described; Direct Numerical Simulation (DNS) using second and high order methods, Direct Simulation Monte-Carlo (DSMC) used for rarefied gases, and triple deck theory used as a validation tool and method for characterising stability. The necessity for high order base-states is justified by the presence of unsteadiness in second order methods, in the form of small amplitude oscillations propagating throughout the domain even at low Reynolds numbers. These unsteady artefacts disappear when using high order methods, as well as the far greater decay and convergence of residuals achieved using these methods. These high order base-states show excellent agreement in the vicinity of separation with triple deck theory which provides confidence in the accuracy of these base-states. Once appropriate base-states are calculated, modal and non-modal stability analysis can be undertaken. Throughout the planar and axisymmetric compression corners studied, an intimate link exists between the separation region and associated shocks through the modal mechanism. At high Reynolds numbers, a stationary unstable mode is identified, taking the form of the well documented separated boundary layer mode. This mode is studied over a wide range of Reynolds numbers characterising the neutral loop of this flow. An analogous mode is found using DSMC that is stable and on the hollow cylinder-flare that becomes unstable at high enough corner angles and Reynolds numbers. Further travelling modes are identified that are modally damped but show the potential to grow downstream out of the domain, which is corroborated using local spatial analysis. These structures are also compared with those of experimental work, showing a similar shape of periodic perturbations triggered in the separation region. Non-modal analysis is performed on unstable planar ramp cases and describes significant short term growth which has the potential to bypass the modal mechanism, even when this modal behaviour is unstable. The large time required for the modal growth to achieve the same non-modal maximum gain suggests this modal behaviour will be bypassed in the transition mechanism. On the stable modal scenarios shown using high angle DSMC base-states, large energy gain is shown at short time suggesting that, in this case even a modally stable case may transition to turbulence.

This thesis examines the interaction between the literary motif of the 'monument poem', and the poetic description (ecphrasis) of material monuments, in the Flavian poet Statius' collection of occasional praise poetry, the Siluae. It explores the close relationship between poetic and material commemorative practices, and the theme of the poet's and patron's aspirations for immortality, within the dynamic and volatile political context of the Domitianic regime. A central concern of this thesis is the interaction between the Siluae's complex poetic programme and the material world with which it engages, and the connection between the cultural capital of the poet and the social capital of which the patron is custodian. Scholarship tends to oscillate between formalist and historicist interpretations of the Siluae, focusing either on the ways in which Statius' poetry can serve to 'mystify' and lend symbolic capital to the material conditions of his patrons or, on the other hand, to focus on the cultural and historical import of the poems, to the detriment of the Siluae's poetic value. The following study focuses on the issues that arise when the traditional tropes of Roman poetic discourse and literary history which Statius uses are reshaped through material referents of the monuments described. This is to be clarified by looking at how 'Statius' monuments' illuminate the connection between poetics and politics in the Siluae. Statius' descriptions of monuments provide opportunities to heavily politicise the aesthetics of the Siluae. Domitian's monuments are used by Statius as analogies for the Siluae's poetic programme, resulting in a deliberate interchange between poet and princeps. This theme is further complicated through Domitian's own complex representation as a poet in Flavian literature, a point often overlooked. Crucial to my reading is the underlying presence of Nero in the Siluae, whose monumental footprint in Rome, and status as an emperor-poet, were both obstacles and opportunities for Domitian. Just as Domitian's monuments work to reshape the memory of Nero in Rome, Statius' Siluae deconstruct Neronian literature and rehabilitate it to suit a Flavian agenda.

Alongside tremendous challenges in infectious diseases, like the rise of antimicrobial resistance and the coronavirus disease pandemic, the 21st century is also witness to the big data revolution, which offers opportunities to design methodology capable of addressing these great challenges. Whilst developing tools there are two competing philosophies of how to gain insight from big data: The modelling approach, where the natural data generating mechanism is approximated by statistical inference, and the algorithmic approach, where general-purpose algorithms are tuned to capture hidden structure in the data for prediction. The aim of the thesis is to contribute existing infectious disease problems, by motivating, designing, and applying the correct big data methodology, whilst facilitating future use through generating applications can be easily re-purposed. I first design a machine learner that can predict the source of Campylobacteriosis 33% more accurately than the previous most commonly used methods. Our method broadens the data input spectrum to captures of whole genomes, which uniquely allows assigning sources to individual samples showing a shift in host affinity of one of the most common lineages of Campylobacter jejuni. Based on the individual prediction of the machine learner,I infer which genetic changes are associated with host specificity by conducting a genome-wide association study. I find fluoroquinolone resistant genes pre-adapting chicken isolates to infection for humans and polyphosphate pathway associated genes to distinguish adaption to chicken and ruminant niche. For the study of COVID-19 risk, I conduct a machine learning prediction of very severe forms of the disease, hospitalisation, and susceptibility, whilst also inferring risk factors for all phenotypes by applying Bayesian model averaging. I re-discover commonly defined risk factors describing socio-economic standing, ill health and ethnicity whilst discovering more novel factors like previous lung injury predisposing very severe COVID-19 and bring order to the wealth of published COVID-19 risk studies. In the closing arguments I give limitations of my work and give recommendations on how the developed tools can be re-applied to make big data research more accessible. I also expand how statistical inference and machine learning prediction can be used in unison to tap into the potential of big data to address the foremost infectious disease challenges of our time.

This thesis explores the durability of elite domination in the three Lebanese municipalities of Zahle, Baalbek, and Tripoli from 2010 to 2019 in the midst of the Syrian migration crisis. It compares how these clientelist actors endeavoured to retain local power after an unprecedented electoral challenge during the 2016 local elections. Since Lebanon's independence in 1943, local politics has been controlled by competing networks of patronage which resiliently maintain their grip on municipal power. Nevertheless, the 2016 local elections revealed an unusual defiance against incumbent clientelist leaders in municipalities like Baalbek and Zahle. Moreover, in Tripoli political independents overthrew the oligarchic elites. These results outlined the frailty of clientelist loyalties in the context of a considerable socioeconomic crisis induced by the Syrian forced displacement into Lebanon. However, only two years later, the dominant clientelist actors in each municipality won the 2018 parliamentary votes in the three case cities. Patronage prevailed, once again. Therefore, the research question framing this project is: "Why did clientelist elites encounter a disruption of their leadership, and then how did they comparatively perpetuate (or recover) their local power in a context of stark material scarcity?" Clientelist domination rests structurally upon three pillars: materialism, symbolism, and collective action. Accordingly, each typology of clientelism features a set of resources which determine their respective strategies of power competition at times of crisis. This research reveals that clientelist actors do compensate or substitute ailing material means with symbolic, coercive or collective tactics of governance. Thus, despite their different profiles, clientelist actors possess comprehensive tools of power endurance supporting their resilience over time. This research uses a mixed-methods approach. It presents original fieldwork qualitative (105 elite interviews and participant observations during the 2018 legislative campaign) and quantitative data (210 survey questionnaires with Lebanese and displaced participants) collected in the three case cities from 2018 to 2019.

What is our Universe made of? Dark matter is a hypothetical non-luminous form of matter whose existence is inferred from a rich collection of astrophysical observations. Its gravitational effects are observed across many scales while its behaviour implies properties not found in known forms of matter and the standard model of particles. The Weakly Interacting Massive Particle (WIMP) is a popular simplified dark matter candidate model that agrees with observations while predicting weak-scale interactions with normal matter. The LZ experiment creates conditions under which the interaction between dark matter and normal matter may be detected. The main detector is capable of discriminating and reconstructing interactions with target nuclei within its unprecedented seven tonne liquid xenon active region. Reducing background interactions from radioactive sources creates a very clean observable space within which WIMP interactions may be found. Lower detector backgrounds improve the experiment's reach to smaller WIMP interaction cross-sections. The LZ experiment is projected to exclude the spin-independent WIMP-nucleon cross-section down to 1.4×10−48cm2 for a 40 GeV WIMP. One of the most important contributions to background interactions comes from decays within the 238U series, but the 232Th and 227Ac series are still relevant. This thesis lays out the first measurement of decays from these series taken during the commissioning period of the experiment when the detector was operated with cold xenon gas. This is the earliest cryogenic measurement of the assembled detector's backgrounds, made possible through the development of a custom trigger. The activity of 222Rn decays was found to be 3.39 ± 0.13 µHz/kg (assuming 7 tonnes liquid Xenon), demonstrating a factor of two suppression from room temperature emanation results. A further suppression in activity is expected with liquid xenon and the engagement of the integrated radon reduction system. The activity of 220Rn, 219Rn and 210Po plate-out (surface accumulation of parent 210Pb during assembly), was also measured and found to be in agreement with projections for the detector.

Two technologies are predicted to enable the third industrial revolution in this beginning of the 21st century: evolutions of mobile communication networks, and artificial intelligence (AI). As 5G is being deployed commercially, and 6G is being investigated by researchers, making full use of the possibilities offered by recent network technologies remains an open challenge. With an unprecedented number of devices connected to one another, new methods are required to design communication systems. On the other hand, AI and its variants (including 'machine learning' - ML, and 'deep learning') have made tremendous progress in recent years, leading the advances in fields ranging from image recognition to natural language processing. Two research directions naturally arise from the connections between those fields, that is, to use communications for machine learning, and to use machine learning to improve communication systems. The former stems from considering that the main training paradigm for ML models still involves performing computations at a single location, with centrally localized data, using supercomputers equipped with sometimes thousands of GPUs. This is problematic in practical situations where data is distributed over many battery-powered devices, with issues both in terms of privacy and efficiency: communication networks are therefore key enablers to deploy distributed machine learning models. We first investigate this aspect by devising an offloading strategy to train machine learning models at the network edge. We then devise a collaborative learning algorithm for highly energy-efficient machine learning models, and show advantages in terms of communication load. On the other hand, AI now appears as a promising alternative to improve communication systems which have so far mostly relied on handcrafted algorithms and approximations of wireless channels dynamics. We demonstrate this by deriving an end-to-end communication system where source and channel coding are jointly learned. Furthermore, both mobile communications and machine learning are ever more energy intensive, creating novel challenges that we aim to tackle by leveraging energy-efficient methods inspired by the workings of biological brains. Using these, we design highly efficient edge AI applications with on-device training, with practical applications in real-world situations.

This thesis explores the philosophical ambiguity and reflexivity of resistance through an investigation of its place. The place of resistance becomes relevant as the intersection of diverse manifestations of resistance and the conceptual reflection through which they come to be explained. Place here denotes a moment of orientation in which reflection and determination coincide and which produces the opposition between resistance and that to which it resists. For my investigation into the structure of this place and its ambiguous entanglement within that to which it resists, I draw on the confrontation between Heideggerian phenomenology and Frankfurt School critical theory and their respective readings of Kant. I ground my argument in an analysis of Kant's critique of amphiboly in the Critique of Pure Reason. In Kant's argumentative strategy to counter the amphibolous use of the understanding, I identify two conflicting directions that find expression in the entwinement of reflection and determination in 'transcendental place'. I contend that Kant's argument relies on an experience of incongruity that remains external to the philosophical argument, and that this pattern is instructive for understanding the place of resistance. I flesh out the nexus of philosophy and politics thus opened up by tracing the persistence and transformation of amphiboly in Heidegger's philosophy and Reiner Schürmann's reading of it. While Heidegger politicises place by turning it into the concrete standpoint from which to oppose modernity, Schürmann reflexively historicises place as the conflictual temporal site of philosophy and politics characterised by a discordant double bind. To further contrast Heidegger's determination of place as concrete, I then turn to the status of abstraction in the constitution of the place of resistance. With reference to Alfred Sohn-Rethel and Moishe Postone, I argue that any opposition conceived on the basis of a concrete standpoint risks obliterating the abstractions that underpin even what presents itself as most concrete. While Sohn-Rethel's affirmation of production remains in thrall to the capitalist metaphysics of labour, Postone's notion of a dynamic concrete-abstract antinomy contributes an important dimension to the conceptual framework for the analysis of the place of resistance. I put Postone's critique of simple opposition to work by analysing antisemitism as a form of opposition that fails to construct a resistant standpoint, arguing that the paranoid mode of thinking driving antisemitism can be understood in terms of amphiboly and that it plays a crucial role in the reproduction of modern capitalist society. In conclusion, and to contrast antisemitism, I bring together Schürmann's and Postone's emphases on the temporal character of modern capitalist domination to briefly sketch an emancipatory orientation of resistance that challenges amphiboly based on a twofold provisional judgement in and on time.

Current computer vision algorithms largely depend on the availability of images labelled by human annotators at very high speed. The mode of production of these annotations strongly resonates with an early experiment conducted in 2007 at Caltech by Fei Fei Li, initiator of ImageNet, one of the most popular visual datasets. In a laboratory, the subjects were asked to describe photographs shown for a few milliseconds and to filter them through a taxonomy. The Caltech experiment is used, in the thesis, to engage with the photographic elaboration of computer vision: the model of vision, the photographic alignments and the micro-temporal rhythm that subtend the modes of production of labelled data and the labour behind it. The written and practice components of the submission elaborate a novel method and document the path towards it. The method has developed in the context of practice-led research in collaboration with The Photographers' Gallery and crystallised into a project, Variations on a Glance, a series of re-enactments based on the Caltech experiment. The original experimental protocol is submitted to several variations, called re-experiments, exploring its potential to produce a time-critical model of vision and collective visual interpretations. The experimental protocol is re-designed iteratively to explore specific configurations of micro-temporal vision and different configurations of collectives of human and non-human participants. The thesis examines the dynamics of these collectives, in particular how they reach consensual interpretation, and how the taxonomic practices of the lab interfere in this process. The contribution of this research is a mapping of the entanglement of computer vision and photography and a method embedded in practice that does not attempt to resolve the differences and tensions between photography and computer vision but provides a device to explore the texture of their relation. The research complements and complicates the recent critiques related to bias and discrimination in machine learning and the exploitative work conditions it relies on. Finally it offers to the photographic institution and its public a mode of intervention into the making of computer vision.

Carbon-based electrodes currently used in the all-vanadium flow battery suffer from loss of electrode material due to the oxidation of C to CO2. Therefore, it is important to develop new electrodes, considering materials that have been used in electrochemical applications, such as the dimensionally stable anodes (DSA). The electrocatalytic characteristics of DSA change with the coating composition and the preparation method and can be adjusted to the electrochemical process. The reversibility of the V(IV)/V(V) reactions and the endurance to the corrosive conditions of the positive electrolyte were the main criteria to select a desirable electrochemical performance. This was evaluated with CV, charge-discharge testing, SEM, EDS, XRD, EIS, and the morphology factor (ϕ). The carbon-based electrodes showed overlapping of the C oxidation reactions in the potential window of the V(IV)/V(V) reactions, and inferior selectivity towards them in the charge-discharge test. The Au and Pt electrodes showed parasitic reactions in the potential window for the V(IV)/V(V) reactions. The Ru DSA electrode showed parasitic reactions and deactivated in the charge-discharge test. The Ir DSA prepared by different methods showed similar electrochemical performance and selectivity towards the V(IV)/V(V) redox reactions, and the main difference was in the damage of the coating after the charge-discharge test. The DSA prepared by the polymeric precursor method suffered the biggest damage with a change in ϕ of 41 %, the DSA prepared by the ionic liquid method had a difference in ϕ of 24 %, and the change of ϕ in the DSA prepared by the template method was 23 %. The DSA prepared with different substrate geometries suffered a bigger change in the ϕ with a 78 % change for the mesh substrate and 54 % change for the micromesh substrate.

The fault location problem has been tackled mainly through impedance-based techniques, the travelling wave principle and more recently by machine learning algorithms. These techniques require both current and voltage measurements. In the case of impedance-based methods they can provide multiples solutions. In the case of the travelling wave approach it usually requires high sampling and synchronized frequency measurements together with sophisticated identification algorithms. Machine learning techniques require training data and re-tuning for different grid topologies. In this work we propose a new fault location method based on the fault's transient intermediate frequency response of the system immediately after a fault occurs. The transient response immediately after the occurrence of a fault is characterized by the travelling wave phenomenon together with intermediate frequencies of oscillation in the range of 5 to 500 kHz. These intermediate frequencies of oscillations are associated with the natural response of the cable/line system to the fault event. Their frequencies of oscillation are dependent on the faulted section and the fault location within that section. The proposed fault location methodology aims to leverage on that dependency, by firstly identifying these intermediate frequencies for different fault location scenarios for a given network. This process is performed offline using a linear time invariant (LTI) representation of the network. To compute this LTI representation, as part of this work an impedance representation in the modal domain is established for cable/line sections, which is able to capture the frequency-dependence and distributed nature of its electrical parameters. The offline methodology identifies these intermediate frequencies for different fault location scenarios, and then proceeds to fit the fault location dependence of each intermediate frequency using a polynomial regression. An online methodology is also proposed to perform the fault location in real time by solving the polynomial regressions computed during the offline methodology using measurements of the intermediate frequencies present in the frequency spectrum of transient signals. The fault location is thus solved by using voltage or current measurements of the fault's transient response at different locations in the network, together with simple signal processing techniques such as the Fast Fourier Transform. The full method is tested with an EMT simulation in PSCAD, using the detailed frequency dependent model for underground cables, together with realistic load models in a low voltage distribution network test system.

The use of micro- and nanostructured surfaces has been explored as a new strategy to enable special surface functions in a variety of areas such as wettability control, anti-biofouling, optical properties, tribology studies, and adhesion resistance. Many manufacturing methods can be applied for producing structured surfaces, including lithographic techniques, femtosecond laser machining or single point diamond turning. However, the cost has often been a critical barrier for the fabrication of functional surfaces on a mass scale. 3D printing techniques have recently emerged as a viable technology due to low material consumption, and high energy efficiency. Among these, Fused Deposition Modelling (FDM) provides a simpler manufacturing process and a more cost-effective method than other prominent 3D printing techniques. Despite the constant improvements in the resolution of this method, FDM still has the disadvantage of excessive extrudate expansion, limiting its application in precision manufacturing of functional structured surfaces. This work aims to develop a novel manufacturing method based on FDM for the fabrication of functional microstructures on thermoplastic polymers in a more cost-effective manner with high throughput on very large scales, which potentially could be applied in the fabrication of bioinspired functional surfaces. For this purpose, the study first assesses the manufacturing accuracy of FDM by measuring the die swell effect of extrudate PLA filaments under varied working conditions in terms of extrusion temperatures (170 ⁰C - 210 ⁰C), printing speeds (10 mm/s - 80 mm/s), layer heights (0.10 mm - 0.40 mm) and nozzle sizes (0.20 mm - 0.40 mm). Different cooling methods are also applied in the printing process to suppress the die swell effect and therefore improve the manufacturing accuracy. By revealing the limitations of conventional theoretical models for predicting filament die swell in FDM process, a numerical model based on the level-set technique is developed to predict FDM extrudate die swell under different extruding conditions. This simulation method is also applied for studying the expansion of patterned structures during the extrusion process. The functions of structured surfaces are demonstrated by enhancing the wettability of a variety of polymers, including polyvinyl chloride (PVC), polyethylene 1000 (PE 1000), polypropylene copolymer (PP), and polytetrafluoroethylene (PFTE). It has been found that materials initially exhibiting common wettability properties (θ ≈ 90°) can exhibit "superhydrophobic" behaviour (θ ˃ 150°).

This study explores the autobiographies of ordinary Britons, written in the middle and later decades of the twentieth century. It draws on a little fewer than 300 of these under-studied documents, reading them for their authors' preoccupations as they sat, writing. Together the autobiographies reveal a wonderful variety of expression. They often take us closer to the people who wrote them. But they also hint at a deeper history. The way people described their lives related not simply to their individual proclivities, or the circumstances in which they wrote, but to their habits of oral storytelling, to their practical skills and routines, and to their knowledge of written genres. Surprisingly, remarkably, the autobiographers' use of speech, practice, and written models suggests broadly comparable features - similar kinds of cultural resource that people used to find some meaning in their lives (Part I). Individual autobiographers drew on these in their own ways and for their own purposes. Yet their access to cultural resources was uneven. Many autobiographers, educated only to a minimum level of competence, found less significance in the written word than they did in oral and practical cultures (Part II). Conversely, authors who mastered written genres tended to see themselves as a cut above the rest, socially and intellectually (Part III). The consequence, especially of different kinds of schooling, was a marked divide in the way autobiographers thought about themselves and the world at large. In this, the autobiographies hint at a profound and lasting separation in British culture: a split in the way people made sense of life; a division persisting at least from the late-nineteenth to the late-twentieth century.

Haptic communication between humans plays a vital role in society. Although this form of communication is ubiquitous at all levels of society and human development, little is known about how synchronized coordination of motion between two persons would lead to higher-order cognitive functions used in haptic communication. During this project, a novel experimental paradigm was developed. In it, participants used their hands to control the rod to collect the coins on the screen. The aim was to investigate the neural and behavioural correlates of emergent haptic communication between paired participants. During the experiment, haptic interactions and neural synchronizations between paired participants were recorded and compared to baseline pairs. Baseline pairs were created from non-interacting single participants. During the Ph.D., two experiments were conducted. The first focused on haptic interactions and characterising the emergent proto-language that participants used to communicate. The second experiment used a novel hyperscanning methodology to isolate the synchronization characteristics of emergent proto-languages and social interactions. The EEG experiment presented here explores the interbrain synchronization between interacting participants systematically. This paradigm allows participants to interact with each other freely. This genuinely novel paradigm allows researchers to study pairs of participants in truly social situations. By comparing pairs with single participants, it has been possible to identify behavioural and neural characteristics related to the emergent proto-language that participants are using as a means of communication. Both of the experiments detailed here describe pioneering work in the study of social interactions and the emergence of protolanguages.

Turbulent transport is a major limiting factor for the energy yield of magnetic- confinement fusion experiments. Turbulence is driven by gradients in the profiles of equilibrium plasma parameters, and it can be regulated by the presence of sheared flows. Direct numerical simulations of turbulent transport are challenging because of the broad range of temporal and spatial scales characterising fusion plasmas. In this thesis, a novel algorithm is developed to include an externally imposed, mean rotational flow shear in linearly implicit, local δf gyrokinetic codes. Unlike the discrete-in-time approach currently followed by a number of codes, it treats flow shear continuously in time, allowing for sparser numerical grids to be used. In some cases, the new algorithm leads to a difference in the level of turbulent transport compared to the discrete-in-time approach. In the prevailing paradigm for plasma turbulence, a unique, stationary turbulent state is associated to given equilibrium plasma parameters. This thesis reports the first instance of bistable turbulence in a fusion plasma with rotational shear. Two distinct states, obtained with identical equilibrium parameters in first- principle gyrokinetic simulations, have turbulent fluxes of particles, momentum and energy that differ by an order of magnitude - with the low-transport state agreeing with experimental observations. The occurrence of the two states is regulated by the competition between the externally imposed, mean rotational flow shear and "zonal" flows generated by the plasma. With small turbulent amplitudes, zonal flows have little impact on transport, and the mean shear causes turbulence to saturate in a low-transport state. With larger amplitudes, the zonal shear can substantially oppose the effect of the mean shear, allowing the system to sustain a high-transport state. This poses a new challenge for research that, so far, has assumed a uniquely defined turbulent state.

This thesis concerns the measurement of freedom, and its value. Specifically, I am concerned with three overarching questions. First, can we measure the extent of an individual's freedom? It had better be that we can, otherwise much ordinary and intuitive talk that we would like to vindicate - say, about free persons being freer than slaves - will turn out to be false or meaningless. Second, in what ways is freedom valuable, and how is this value measured? It matters, for example, whether freedom is valuable only insofar as it enables us to pursue specific ends we happen to value for independent reasons, or whether it is also valuable in itself. In the latter case, but not the former, more freedom will always (other things being equal) be better. Likewise, it's important to get clarity on what ends, exactly, freedom is especially instrumentally valuable in promoting, since this goes to how much we should care about it. And finally, the liberal political tradition asserts that there is a special sphere of personal choices within which individuals should be free to do as they please, for reasons over and above the value of freedom itself. Now, can we measure the extent to which states and individuals respect individual liberties, and can we weigh the importance of respecting liberty against competing values like social welfare? To answer these questions, I follow the axiomatic tradition of social choice theory, and I develop several novel measures of freedom - including the first measure of freedom ever proposed that is sensitive to how modally robustly our options are available - a novel approach to measuring the diversity of an opportunity set, and I develop an account of the value of freedom according to which an important part of why freedom matters is that it enables us to improve our preferences through learning, and I construct ways of measuring the value of freedom along this dimension of value. Finally, in later chapters, I provide a representation result for a measure of illiberalism, i.e. of the degree to which states fail to respect the rights of their citizens, and I extend this result to provide a characterization of moral theories that express concern for respecting the rights of others. The final chapter closes by discussing how this concern may be weighed against concerns for social welfare.

This research investigates streetspace allocation metrics for all streets in London providing quantitative evidence about a key parameter of street design citywide. A new methodology to quantify streetspace allocation is introduced using a geocomputational approach that allows both the processing of high-resolution topographic data over a large geographic extent and enables replicability for other cities. The correlation between streetspace allocation metrics and street network centrality at distinct scales is investigated across different geographic areas. These variables are then examined using cluster analysis to identify a typology of streets based on streetspace allocation and centrality. The results provide the framework for a design scenario study of inner London applying shortest-path analysis under an active travel prioritisation perspective. Streetspace statistics for London confirm the predominance of space allocated for vehicular transport over pedestrian uses. Most streets display standard "residential" street metrics, coinciding with traditional street classification schemes. Also, this serves to demonstrate quantitatively the spatially efficient organisation of the London street system with few wider distributors and many narrower local streets. In addition, through the combined examination of the streets' allocation and configurational metrics, it is possible to identify a new sub-type of local streets. The spatial arrangement of the streets segments types follows a centre-periphery pattern: wider and higher centrality streets are clustered at the city centre and show relative larger streetspace designated to pedestrians, corresponding with higher levels of estimated activity. On a prescriptive streetspace model of Inner London, the streetspace allocation of critical pathways is modified to illustrate how strategic scale street properties affect and are affected by design scale street parameters. The fine-grain physical metrics analysed here, not only can be useful to tackle a wide range of contemporary street related questions from urban environmental quality to the adoption of new technologies but also offer alternative analytical methods for street research, planning and design.

This work explores new ways of preparing reactive polymer-based nanoparticles prepared through polymerisation-induced self-assembly (PISA). PISA is an emerging and advantageous method to prepare well-defined diblock copolymer nanoparticles but the literature up until now has only demonstrated few examples of incorporating reactive functional groups. This work centres on the 2,3,4,5,6-pentafluorobenzyl (PFB) moiety - an underexplored reactive functionality - in the monomer 2,3,4,5,6-pentafluorobenzyl methacrylate (PFBMA). In order to explore the reactivity of the PFB group the monomer was reacted with sodium azide to give, in quantitative conversion, the 4-azido-2,3,5,6-tetrafluorobenzyl-functional species, a previously unreported compound. This species was further transformed into amine, amide, 1,2,3-triazole and amidine. Next, PFBMA was used in PISA formulations with poly[poly(ethylene glycol) methyl ether methacrylate] (pPEGMA) as macromolecular chain transfer agent. pPEGMApPFBMA nano-objects presented spheres, worms, or vesicles morphologies with examples of worms samples forming free-standing gels that underwent a reversible degelation and morphological transition to spheres upon heating. This work demonstrated that PFBMA is as versatile as the commonly used benzyl methacrylate. Significantly, however, the use of PFBMA allowed the inclusion of reactive functionality into the nanoparticle cores. This is novel as most literature-known examples of reactive PISA particles feature corona-reactive species. Herein, the reactivity of the PFB group towards nucleophilic aromatic substitution with thiols was exploited and several key advantages of the novel reactive PISA particles are highlighted: (i) the use of 1,8-octanedithiol to cross-link the cores prevented particles disintegration in non-selective solvents and resulted in the loss of the thermo-responsive behaviour; (ii) using polar thiols, particles underwent a order-disorder sphere-to-unimer transition, loosing their well defined character; (iii) large thiols resulted in size increase or order-order sphere-to-worm transition including gelation of worms. This latter case represents the first example of a chemically triggered morphological transition following an increase of the packing parameter. It is anticipated that the understanding gained in this work will contribute to the future development of reactive "universal nanoparticle precursors" that can be easily modified into functional nanoparticles with predictable size, shape, and behaviour on demand.

This thesis focuses on how the Precautionary Principle, a general principle of international law, can strengthen the protection of Indigenous peoples' rights, when States' actions may affect their lands and territories. This brings into focus the importance of Indigenous peoples' traditional knowledge, a system of knowledge tied to their lands, systematically developed by relying on observations, practices and experiences, which is able to predict outcomes. This knowledge represents an integral understanding of Indigenous peoples' traditional way of life, providing the 'best information available' about customs, values, activities, their special relationship with their environment and even the existence of particular risks and uncertainties, making it vital when it comes to the adoption of informed decisions and the full understanding of the possible consequences of a determined action on their livelihood. This thesis shows that the precautionary principle has been influenced by a narrow understanding of one of its elements, 'scientific uncertainty', insufficient to effectively address potential threats that originate in human activities, like the extraction of natural resources, and seriously disturb the complex web of relationships and dynamics in determined ecosystem. This makes it necessary to embrace other sources of knowledge, in order to effectively avoid the materialisation of these potential impacts. It argues that traditional knowledge is recognised in international law as a source of scientific knowledge in a broad sense, and thus it should be integrated into the element of 'scientific uncertainty', providing a different understanding of risks, of 'what is known/unknown, what is uncertain and what is controversial. Moreover, as a scientific basis for the adoption of precautionary action, the integration of traditional knowledge expands the application of the precautionary principle to deal with potential threats of non-negligible harm to Indigenous peoples' traditional way of life. From a procedural point of view, the integration of this traditional knowledge requires the existence of participatory mechanisms of public participation. In this case, the two participatory mechanisms that are at the front and centre of this discussion are the States' duty to consult Indigenous peoples, with the objective to obtain their free, prior and informed consent, when they may be affected by certain measures; and the implementation of impact assessments. In this sense, I argue that both mechanisms are precautionary measures, being triggered when there are 'reasonable grounds for concern' about possibly non-negligible harm to Indigenous peoples' rights. This means that States must not only conduct consultations/impact assessments but are also under the duty to avoid the materialisation of potentially serious harm to Indigenous peoples, which implies the need to consider their traditional knowledge and accommodate their concerns, at the risk of failing to comply with the precautionary principle. In addition, following a precautionary logic of 'the higher the risk the greater the need for precaution', States must reach an agreement or obtain Indigenous peoples' consent in situations where the potential impacts on their traditional way of life are of a more substantive nature. This shows that the precautionary principle strengthens the protection of Indigenous peoples' rights.

Governments are increasingly implementing public sector innovation laboratories (PSI-Labs) to support public innovation. Although most of these Labs use design approaches to drive innovation, previous research shows that the impact of such approaches is limited. This is due to the lack of integration of design with policy processes and power structures, the weak legitimacy that design knowledge has as a source of policy evidence and the incapacity of design to turn the public sector's politico-organisational factors into matters of design. This design research project aimed to further understand the value of design in Public Sector Innovation and to clarify the role it can play in the different levels of government transformation when introduced and integrated into the politico-organisational context of a new PSI-Lab. Through a process of 'action research through design', this project shows how design was used as a core capability to create, implement and operate a new PSI-Lab at the highest levels of central government. Through the reflections from the process of practice, strengthened by a qualitative study of the case, this thesis explains how design was effectively introduced and integrated into the politicoorganisational context and policy cycle of this new PSI-Lab, helping to increasing its impact and scale. This research demonstrates that integrating design into the politico-organisational context of PSI is possible when: (1) design is practised as an attitude of turning these factors into matters of design; (2) the practices of design and politics are intertwined in building legitimacy and social capital; and (3) policycycles are understood as adaptive and dynamic such that policy design and implementation are practised as two interdependent and experimental activities. Once the barriers of integration had been overcome, the study shows that this specific PSI-Lab was able to scale design as an innovation capability across a broader spectrum of the public sector, helping to reposition design in public sector innovation research and practice as an open, flexible and integrative approach to deal with public problems. Finally, this study offers a body of knowledge from practice on how to design and implement a PSI-Lab in a large public sector system.

Large herbivores are an important natural resource for humans, but a high proportion of these species are classified as globally endangered. Current declines of these species are mainly caused by unsustainable harvesting and land use loss or conversion. These processes frequently increase herbivore mortality rates, and alter a species' behaviour and local distributions. Because of their importance, humans frequently attempt to manage large herbivores to assure their sustainable use and effective protection. Nevertheless, these actions usually focus on managing and monitoring target species, neglecting possible side effect on other species or components of the environment. In this thesis, I focus on studying both environmental and anthropogenic effects on a large herbivore community that inhabits a seasonal, human-modified arid savanna. Firstly, I propose a method to automate monitoring body mass variation of large herbivores over time and use this to explore seasonal mass changes. The method permitted continual body mass monitor of three species over an 18-month period. Animals lost weight during the dry season and gained weight during the wet season, though interesting sex-specific variation in mass change phenology occurred. Frequency of precipitation was the main predictor of annual mass variation. Such work has the potential to inform on the potential impacts of changing future climatic regimes on large herbivores. Secondly, I map the spatial abundance of 16 species using Distance Sampling and Density Surface Models. My results suggest that this highly diverse community of large herbivores can coexist in African savannas because they spatially differentiate their vegetation and water preferences within and between seasons. I found that both target and non-target hunted species avoided human activities at my study site, suggesting that management activities in the area are perceived as a threat, and that hunting has impacts beyond target species. Finally, I use Structural Equation Modelling to assess the simultaneous effects of prescribed fire on tick abundance, grass volume and large herbivore abundance. My results show that ticks decline in number during the first 12 months after burning, but then rapidly return to previous densities. Burning vegetation has a longer-lasting effect on grasslands, and grass volume only returned to pre-burn levels after three to four years. My findings highlight the need for conservation managers to monitor the effects of target actions to assess impacts on both focal and non-focal organisms as well as on the health of the wider ecosystem. Here I have presented a tool to monitor the day-to-day condition of the larger wild ungulates of African savanna systems, which could inform management activities (such as relocating animals, provision of additional resource etc) based on realtime, empirical data. I have demonstrated that relatively inexpensive ground surveys can provide consistent, regular indices of the abundance of ungulate species. The role of human activities in altering herbivore use of the landscape indicates how reducing such activity might actually increase savanna ungulate carrying capacity. These findings, along with the more detailed understanding of the role of fires on the vegetation-grazer-parasite system that I present, should aid conservation managers and researchers in conserving such species, under both current and future scenarios.

In this thesis a novel method for quality control of the cooling half-rings in the pixel endcaps is described. The half rings are used to provide the cooling for pixel modules in the new Inner Tracker of the ATLAS experiment that will operate during the runs of the High Luminosity LHC. In the quality control, they are analysed using infrared measurements during cooling processes. Defective half rings can be found and rejected during the serial production. The sensitivity of this method is shown for small ring pieces with known implanted defects, as well as for a half ring prototype. The measurement of the ttH production cross section with H! bb decays for the full Run II data set, corresponding to an integrated luminosity of 139.0 fb-1 at a centre of mass energy of p s = 13 TeV, is discussed. The signal strength with respect to the Standard Model expectation is measured to be: (ttH = 0.84(+0.49-0.44) (+0.21-0.21 stat.)(+0.44 -0.39 sys.) , which corresponds to an observed (expected) significance of 1.9 sigma (2.2sigma). Additionally, a new analysis strategy to measure the CP-nature of the t-H-coupling in the ttH production is described. This analysis is expected to exclude the purely CP-odd scenario at almost 95% confidence level.

Scholars have highlighted the role of income distribution as a fundamental factor to understand consumption, health, adoption of technologies, social cohesion, democratic stability, and long term economic performance, among other phenomena. Moreover, reducing income inequality was included in the United Nations’ Sustainable Development Goals. However, less attention has been put on the link between differences in household incomes and other pressing sustainability challenges, such as those that depend on massive adoption of new technologies within network industry sectors: telecommunications, waste management, transport, energy and water. Does existent income inequality translate into adoption gaps within these sectors? Is inequality an obstacle for advancing the vision of an inter-connected and more sustainable world? This research explores these questions through seven stand-alone papers, which focus on adoption of broadband internet, municipal recycling and railway passenger transport. Part I of the thesis includes four papers based on publicly available data from traditional mainstream sources. Chapter 1 provides a systematic map of the peer-reviewed literature on the link between income inequality and adoption of the three selected network technologies. Chapter 2 looks at country-level panel data from OECD countries. Chapter 3 analyses cross-sectional data in a broader world-wide sample. Chapter 4 compares borough-level recycling and income distribution in two European cities: London (United Kingdom) and Barcelona (Spain). Part II provides an in-depth analysis of two South American metropolitan areas: Santiago (Chile) and Medellín (Colombia). This part includes three papers (chapters 5, 6 and 7), each one focusing on a specific sector, that employ mixed-methods based on fieldwork conducted in both cities. The main contributions of the thesis are new evidence on the negative effect of income inequality on network technology adoption, and a discussion of the role of formal and informal institutions in this relationship.

Within this thesis we examine how security verification and validation (V&V) is practiced, supported, and perceived, within small to medium-sized software enterprises (SMEs) based within the United Kingdom. Further, we show how the activities associated with this complex socio-technical practice are influenced by the socio-technical characteristics of such organisations - specifically, by determining whether organisational information security culture influences security V&V practice and by establishing whether organisations can possess a mature security V&V practice without a correspondingly mature software V&V practice. Acknowledging the social-technical nature of V&V, this study is guided and framed through the adoption of the Socio-Technical Interaction Network (STIN) framework. Further, and given that the organisational and technological contexts of this thesis are relatively underexplored, we employ a mixed methods, sequential explanatory design approach towards data collection and analysis. This involved an initial, predominately quantitative web-based survey, with the resultant themes identified being further explored during the subsequent, qualitative semi-structured interviews. This study contributes to existing empirical software engineering research by highlighting the socio-technical realities surrounding security V&V practice within UKbased software SMEs, hitherto an empirical unknown. Our findings also show that such practice is influenced by an organisation's information security culture and that a relationship between an organisation's software and security V&V practices exists. Notably, we find that software SMEs need to improve their security V&V practice. Whilst this may have been inferred by the increasing number of vulnerabilities being reported, our study highlights that any improvement must encompass more than the technical (e.g. the adoption of specific techniques and tools), and actively include the social (e.g. addressing employee relationships, raising awareness levels and proactively acknowledging the drivers, and influences, which exist both internal and external to an organisation), since all these aspects are intertwined. Further, this study contributes to the existing STIN literature by showing its suitability for studying security V&V.

This PhD thesis uses aspects of a criminology framework to examine the extent to which research on small arms and light weapons (SALW) undertaken to support international policy is relevant and replicable beyond its immediate field of practice. Using a sample of six primarily field-researchbased publications, I examine whether this research generated a greater understanding of the most problematic uses and users of SALW, and the role of these weapons as instruments of violence. With respect to uses, the application of public health and mixed social science methods has helped to reduce knowledge gaps on the effects of SALW in developing and post-conflict societies. Estimates of the costs of violence in developing countries demonstrated the instrumentality of SALW—i.e. the more serious societal impacts of firearm violence than those of violence involving other instruments. SALW research on users contributed to expanding the agenda from an initial focus on international trafficking to supply insurgent groups to a more comprehensive examination of the patterns of SALW procurement, management, control, and use among a broad range of actors able to contest the state’s monopoly of coercive force. My work on the instruments of violence contributed to an increasingly precise understanding of the most problematic types of SALW held by criminal, terrorist, and non-state armed groups in Africa and Europe. Finally, replicating field-based black-market price-monitoring techniques in conflict areas showed that ammunition prices and war-related fatalities can be strongly correlated, and provides an important lead for further examining the accessibility thesis—i.e. the link between SALW availability and levels of violence. The present thesis provides several suggestions for moving the field of practice forward. Firstly, there is a need to consolidate the lessons learned from SALW researchers’ extensive use of social science methods—including surveying—in post-conflict situations, and to analyse their implications for the measurement of SALW availability and the incidence of violence more broadly. Secondly, SALW researchers need to engage in scientifically robust evaluations of the impact of the most novel interventions, which would represent significant contributions to both SALW policy and academic research into gun violence. Finally, various streams of SALW research have highlighted the importance of ammunition supply in sustaining conflict and violence, a subject so far largely overlooked by those researching gun violence in developed countries. More expansive inquiry into ammunition flows and their relationship to violence has the potential to represent a major contribution to academic research into gun violence.

Ultrasonic arrays are used in nondestructive testing for a wide range of inspections. The Full Matrix Capture (FMC) acquisition technique allows the capture of all the information possibly measurable by the probe. The Total Focusing Method (TFM) forms images by post-processing the FMC data, superseding conventional phased array techniques. By exploiting the wave mode conversions and the internal wave reflections in the specimen, multiple ultrasonic images may be formed, an approach termed multi-view imaging. Multi-view imaging increases the chance of obtaining a high response from a hypothetical defect in the specimen by considering multiple insonification paths. This thesis investigates several areas relevant to multi-view imaging and defect characterisation. The defect response varies greatly with several variables including its shape, location, orientation, and the inspection set-up. As a consequence, the defect may have a strong response in a view, but be invisible in another one. A fast ultrasonic model which predicts the defect response in any view is introduced to help design sensitive inspections. The model is also used to build a database of reference scatterers against which the unknown defect is compared to characterise it. This technique makes it possible to determine the approximate shape, size and orientation of flaws which are too small to be well resolved on a TFM image. The structural noise ultimately limits the detectability in ultrasonic images by creating a speckle pattern obscuring the defect. The peak amplitude of the speckle is a quantity relevant to the calculation of probabilities of detection and false alarm, but has been little studied in the literature. This thesis pursues its analysis. Delay-and-sum is a widespread imaging approach. The presence of wall reflections may significantly degrade the image by adding artefacts. Replacing the summation by the median, a novel approach, is shown to efficiently suppress these artefacts.

Ultimate lighting sources for general illumination are monolithic on-chip white light emitting diodes (LEDs) containing multiple colour emissions, either red-green-blue or blue-yellow, but without involving any yellow phosphors. It is highly likely that current white LEDs fabricated by using a “blue LED+ yellow phosphors” approach will be eventually replaced by monolithic on-chip white LEDs. One of the direct routes for the fabrication of monolithic white LEDs is to utilize InGaN quantum wells (QWs) with different emission wavelengths as an active region, which will involve a number of fundamental issues, such as the design of an active region, carrier transport, etc. So far, these fundamental issues have not been understood. In this work, a systematic simulation study on these challenging issues has been carried out, achieving a full understanding of these issues and thus leading to the design of optimized white LED structures on (11-22) semi-polar substrates by taking the major advantages of semi-polar LEDs in comparison with their c-plane counterparts. Finally, the monolithic on-chip white LED epiwafer based on these designs have been successfully grown on our well-established (11-22) GaN templates with a step-change in crystal quality. Detailed device characterization has been performed on these LEDs, validating these approaches and designs. The design of dual-colour (11-22) semi-polar LEDs aiming at white LEDs and their carrier transport issues have been systemically studied by using one-dimensional drift-diffusion simulations. Due to the much heavier effective mass of holes than that of electrons and also the much larger activation energy of p-GaN than n-GaN, the distribution of injected carrier (mainly holes) is extremely uneven during LED operation. Furthermore, the residual polarization of semi-polar LEDs makes the case even more complicated. Based on a systematic study, carrier transport issues for (11-22) semi-polar white LEDs and their c-plane counterparts have been fully understood, demonstrating their major differences. In addition a novel structure utilizing an extra thin GaN spacer prior to the growth of blue InGaN quantum well, has been design to effectively improve hole transportation and a dual-colour emission LED has been achieved. A tri-chromatic emission has been subsequently designed by further optimizing two key factors, indium content in InGaN quantum wells and barrier thicknesses. In order to validate our simulation results dual-color emission LEDs have been grown on our high quality (11-22) semi-polar GaN templates. Simulations have agreed very well with experimental results demonstrating that both the growth order of the yellow and blue InGaN quantum wells and the growth of a thin GaN spacer are of vital importance. A different approach has been developed, leading to the growth and then the fabrication of monolithically integrated white light LEDs on (11-22) semi-polar GaN template. In this approach, an electrically injected semi-polar blue LED is firstly grown, followed by a yellow multiple quantum well structure as a down conversion layer. This forms a white LED. For the first time, a systematic and comprehensive study on optical polarization properties has been conducted on (11-22) semi-polar LEDs with a wide spectral region from blue to yellow as a function of indium concentration and injection current. Fundamental understanding of the polarization properties and the emission mechanisms of (11-22) semi-polar LEDs has been achieved. Detailed polarization dependent electroluminescence measurements have demonstrated that both indium content and current injection play crucial roles in the optical polarization properties of (11-22) semi-polar LEDs.

This study considers the politics of epithets from the start of the imperial crisis in 1763 until the Constitutional Convention in 1787. More than mere insults, epithets were defined in this period as appellations or titles and were used to describe a person’s qualities or attributes. Despite the importance of these identity terms as the ideals that people most valued in their neighbours, early Americanists do not centre epithets. Historians focus on individual terms – “whig,” “American,” and “republican” – but these labels have not been brought together into a conceptual history of epithets. When these terms are examined together, this thesis argues that the partisans, the opponents of British rule, invented many of the words to discuss who they were, build bonds of belonging amongst their supporters, and identify their internal and external enemies in the Revolutionary period. In attempting to form a sense of themselves and others in the midst of such a divisive event, the partisans transformed the terms of their British colonial past, labels that emerged over the course of the seventeenth and eighteenth centuries as the colonists started to envision themselves as distinctive but equal to Britain, and reformed them into epithets that they used to demonstrate the virtues of the United States of America, and determine the ideals that inhabitants were meant to live by in a new nation. This process was far from uncontested though. Rather than emerging separately, the partisans developed their epithets in conversation and opposition to several rival terms that emerged in a war over words with their enemies, including Britons in the metropole, a significant number of native peoples in the Ohio river valley, and persons disaffected to the cause in Virginia. Since the partisans invented these terms in such a conflicted environment, they argued that only those people who showed merit were worthy of using epithets. The politics of epithets was the politics of merit. Since merit was a contested concept, America and Britain’s inhabitants struggled over who merited epithets and constantly changed the guidelines over who deserved to use these terms. This contest over words had Janus-faced outcomes. It allowed the “people out of doors,” including poorer whites, women, and black persons, to claim rights and belonging as meritorious “citizens.” Yet the political elites, those “within doors,” who were instrumental in manufacturing these fighting words, ensured that only the chosen few, especially white men, could call themselves “Americans.” The origins of much social inequality, over who had the same status, in the early United States was therefore partly born from a seemingly egalitarian ideal: a society where epithets were only given to those who deserved them.

In order to act, plan, and achieve goals, people must learn about their environment and the outcome of possible actions. One reason for human successes in developing new theories and strategies when confronted with new problems is that people are not passive observers. Indeed, children ask informative questions and can adapt their strategies when inquiring about things they don’t know. In this thesis, I aim to understand how people self-direct their learning across multiple tasks, when trying to achieve goals within their environment. Chapter 2 presents experiments designed to better understand people’s exploration and reward maximising strategies across a sequence of tasks. Through these experiments I examine how the environment, specifically the availability of information, and the prior knowledge of participants, affect their exploratory strategies. To study participant strategies I develop a general framework that considers both Bayesian approaches as well as a range of simpler heuristic strategies to approach the problem of goal-directed exploration (Chapter 3). This framework aims to explain the variation of participant strategies in terms of different underlying cognitive mechanisms that guide exploration. One of the benefits of a general framework is the ability to capture a diverse set of behaviours within a continuous parameter space. I focus on the problem of understanding the differences between participants by leveraging this shared psychological space. Specific families of strategies emerge from the behaviour of participants, highlighting the importance of studying individual di↵erences to better understand cognitive mechanisms (Chapter 4). In Chapter 5, I analyse the experimental data from Wu et al. (2018) that considers similar phenomena concerning human exploration, with a specific focus on people’s ability to generalise to guide their search. My analysis shows that our general framework offers a more compelling explanation for participant behaviour than the model they present, while again highlighting the importance of looking at individual differences. From these model based analysis we find that people are able to adapt to the structure of their environment, and are guided by local uncertainty rather than global uncertainty during exploration. Finally, Chapter 6 looks at participants’ behaviour when learning across a sequence of tasks when the underlying problem structures may change. How do people learn in a changing world? I show that the theory of inference by sampling can help explain distinct phenomena relating to the dynamics of learning across tasks. Our models are able to explain people’s ability to progress across tasks when they share structural similarities, their ability to adapt to change, but also specific contexts where participants are continuously unable to realise the world has changed.

The work presented in this thesis aims at the development and validation of a wavefront sensor concept for adaptive optics (AO) called the pupil-modulated Point-Diffraction Interferometer (m-PDI). The m-PDI belongs to a broader family of wavefront sensors called Point-Diffraction Interferometers (PDIs), which make use of a small pinhole to filter a portion of the incoming light, hence generating a reference beam. This allows them to perform wavefront sensing on temporally incoherent light, such as natural guide stars in the context of astronomical AO. Due to their high sensitivity, PDIs are being developed to address several difficult problems in AO, namely measuring quasi-static aberrations to a high degree of accuracy, the cophasing of segmented apertures, and reaching a high correction regime known as extreme AO. But despite their advantages, they remain limited by their narrow chromatic range, around ∆λ = 2% relative to central bandwidth, and short dynamic range, generally of ±π/2. The purpose of developing the m-PDI is to explore whether this new concept has any ad- vantages regarding these limitations. Indeed, we find that the m-PDI has a maximum chromatic bandwidth of 66% relative to the central wavelength and a dynamic range at least 4 times larger than that of other PDIs. Although the m-PDI concept had been proposed previously, it had not been explored to the extent reached in this manuscript. This thesis presents an initial investigation into the m-PDI, beginning with the development of the theory. Here the theoretical framework is laid out to explain how interference fringes are modulated by the wavefront, how to then demodulate the propa- gating electric field’s phase and then finally how to measure the signal-to-noise ratio (SNR). After building analytical and numerical models, a prototype is designed, built and characterised using CHOUGH, a high-order AO testbed in the lab. This incarnation of the m-PDI is called the Calibration & Alignment Wavefront Sensor (CAWS). The characterisation of the CAWS shows two things. The first one is that the CAWS’ response is approximately flat across its spatial frequency domain. The second one is that its dynamic range decreases at higher frequencies, suggesting that it depends, amongst other things, on the wavefront’s slope. In order to prove that m-PDIs can be used for AO, a control loop is closed using the CAWS and CHOUGH’s deformable mirror, with both monochromatic and broadband light. The results show that the final Strehl ratio increases from 0.2 to 0.66, at a wavelength of 633 nm. The difference in residual aberrations seen separately by the imaging camera and by the CAWS is about 20 nm RMS. This is explained by non-common path aberrations and low order aberrations which are invisible to the CAWS. Finally, the instrument was tested on the CANARY AO bench at the William Herschel Telescope. The CAWS was successful at characterising the quasi- static aberrations of the system and at demodulating the phase of wavefronts produced with the deformable mirror. When demodulating on-sky residual aberrations at the back of CANARY’s single-conjugate AO loop, the SNR remained too low for effective wavefront demodulation, only sporadically in- creasing above 1. These results are not discouraging as the CAWS was only a first prototype and CANARY is not a high-order system, reaching a Strehl ratio of around 0.5% at 675 nm. The lessons and improvements for future de- signs are to increase the diameter of the instrument’s pinhole by at least twice, and deliver it a higher Strehl ratio by moving towards longer wavelengths and employing a higher order AO system.

Kidney disease is a global health concern with high mortality for which effective therapies are lacking. Accumulating evidence has identified excessive mitochondrial fragmentation causing mitochondrial dysfunction as a central pathologic feature of kidney disease. Recently, the NAD+-dependent lysine desuccinylase/demalonylase sirtuin 5 (SIRT5) has emerged as a key regulator of mitochondrial form and function but its role in the kidney is unknown. Immunostaining of normal human kidneys showed increased expression of SIRT5 in mitochondria-rich tubules. Subsequent immunostaining of murine kidneys from an ischaemic renal injury (IRI) (AKI model) and folic acid nephritis (FAN) (CKD model) model revealed that IRI increased SIRT5 expression, while FAN decreased expression. An in vitro oxygen/nutrient-deprivation (OND) model was developed in human proximal tubular epithelial cells (hPTECs) to mimic in vivo renal ischaemia. OND increased SIRT5 expression in hPTECs. SIRT5 was depleted (by RNAi) and hPTECs were exposed to OND followed by assessment of mitochondrial form and function using confocal/ transmission electron microscopy, Seahorse, FACS and WB. SIRT5 depletion impaired cellular energy metabolism, disrupted mitochondrial fission/fusion dynamics, induced mitochondrial fragmentation and enhanced mitophagy. This effect was exacerbated by OND. SIRT5 depletion also increased mitochondrial swelling and decreased mitochondrial respiration after OND. Sirt5-/- and wild-type mice underwent IRI surgery (40min) (AKI model), or received a single injection of with folic acid (240μg/g BW) or vehicle control (CKD model). IRI and FAN kidneys were harvested after 24h and 14d, respectively, and analysed by qPCR, IHC and WB. Sirt5-/- mice were protected from IRI and showed mildly aggravated injury in the FAN model. Analysis of renal Sirt5-/- mitochondria revealed a reduction in complex II activity, a central driver of reperfusion injury in IRI, hinting that this may be one potential mechanism that alleviates IRI. Mitochondrial function has been shown to be impaired in Sirt5-/- kidneys suggesting that this may have exacerbated FAN. Taken together, the data showed that SIRT5 is an ischaemia-inducible enzyme in murine kidneys and hPTECs. This together with the observation that SIRT5 depletion caused mitochondrial dysfunction in vitro in hPTECs as well as in vivo in murine kidney suggested that the increase in SIRT5 levels is aimed at enhancing mitochondrial function. Loss of SIRT5 alleviated IRI and exacerbated FAN in mice in vivo indicating that impaired mitochondrial function may on one hand reduce acute ischaemic injury, and on the other, aggravate chronic nephrotoxic injury.

The aerospace industry is investigating more electrical aircraft since the early 90's. This electrification process in the aerospace industry look at replacing more and more hydraulic and mechanical actuator into electric and electromagnetic ones. The surge in electrical components in the aircraft leads to an increase in electric power consumption. For this reason there is a growing interest into the design of generator for aero-engines. Generator for turbofan are usually attached via a gear box to the high pressure shaft of the turbine. But in the last 20 years, several studies have tried to use electrical generators directly in-line with the turbine's shaft. This raised some challenges in terms of electrical machine performance, heat management as well as finding the good typologies to use for such a task. In order to reduce even further the use of mechanical components in the aero-engine one could use an in-line electrical machine as an actuator for the starting of the turbine. However In this study will look in a different way to use an electrical generator. In addition to generate electrical power it will be able to work as actuator to reducing the impact of the vibration on the shaft. For this task the machine will be using one single winding to perform both functions. A model of the bearingless machine will be set as well as the expressions giving the forces and the torque generated by the bearingless machine. A control system will be set up for that machine based and the model previously found. Furthermore the possibility to reduce the vibration of the shaft using the force generated by the bearingless machine will be investigated. Looking not only at the reduction of the vibration but also the possibility of avoiding critical speeds of the shaft. Different strategies to fulfill that objectives will be proposed. Those strategies will be tested on a common test bench build in collaboration with 3 other PhD students.