This thesis presents both a computational and experimental investigation into light propagation in hot alkali-metal vapours, with a particular focus on utilizing the Faraday effect for practical applications. A model to calculate various spectra for a weak-probe laser beam in an atomic medium with an applied axial magnetic field is presented. A computer program (ElecSus) was developed which implements this model efficiently. Using ElecSus we design optical devices such as Faraday filters and laser frequency stabilizing references. The design of Faraday filters utilizing compact vapour cells is shown, along with excellent agreement with experiment. The importance of including the effect of self broadening in the model is shown for these short path length vapour cells. Also, a Faraday filter is presented that can potentially be used for quantum optics experiments on the caesium D$_1$ line (894~nm). The filter displays the highest ratio of transmission to equivalent noise bandwidth to date for a linear Faraday filter, demonstrating the power of computerized optimization for this application. Furthermore, a Faraday filter is presented for use as an intra-cavity element in an external-cavity diode laser. A proof-of-principle experiment is demonstrated which shows that using a short external cavity with the Faraday filter eliminates mode-hops. Experimentally and theoretically the Faraday effect is investigated in large magnetic fields where alkali-metal atoms enter the hyperfine Paschen-Back regime. This hyperfine Paschen-Back Faraday effect is shown to allow a direct measure of the refractive indices for left and right circular polarized light. Furthermore, fitting the weak-probe spectra using ElecSus is found to give measures of the magnetic field with a fractional precision of the order of $10^{-4}$. In addition we study slow-light pulse propagation in a high density rubidium vapour, showing that our theoretical model for the electric susceptibility is valid for short pulses as well as continuous-wave light. This shows that the model is accurate for predicting weak-probe pulse propagation.

Ultraluminous X-ray sources (ULXs) are extra-galactic, non-nuclear point sources, with X-ray luminosities brighter than 10^39 erg s^-1, in excess of the Eddington limit for 10 M_sun black holes. Recent results indicate that the majority of ULXs are stellar remnant black holes accreting material at or above the Eddington rate, rather than sub-Eddington accretion onto intermediate mass black holes. However, precisely how these ULXs accrete material at a super-Eddington rate remains an open question. This thesis focuses on the nature of these system as well as their environments, and attempts to explain physically how the sources operate in this super-critical accretion regime. This work begins with a study of the X-ray spectra of ULXs in very nearby galaxies (D < 5 Mpc). A range of physical models is used to explain the ULX spectra and to interpret the results physically. The outcomes consistently suggest that ULXs are stellar remnant black holes accreting material at or above the Eddington rate. It is demonstrated that the hard spectral component is consistent with emission from the inner radius of an advection-dominated slim accretion disc; the mass of black holes powering ULXs can be constrained from this hard emission, falling in the regime of stellar-mass black hole (~3 - 30 M_sun). Assuming that the soft spectral component represents soft thermal emission from an optically-thick outflowing wind, the size of the wind is constrained to be between ~10^4 – 10^6 R_g. We further explore the nature of ULXs by studying the X-ray spectral evolution of the individual source Holmberg IX X-1 with observed source luminosity. We find that the spectra tend to evolve from relatively flat or two-component spectra in the medium energy band, at lower luminosities, to a spectrum that is distinctly curved and disc-like at the highest luminosities. This spectral variability is consistent with the prediction of super-Eddington accretion models, in which the outflowing wind is expected to be launched from within the photospheric radius; the increase in accretion rate causes the more powerful wind to scatter a higher fraction of hard photons into the line of sight, while those that survive the passage through the wind will be Compton down-scattered to lower energies; these increase and soften the hard spectral component, resulting in a disc-like spectrum peaking at lower energy than the hard component seen at lower luminosity. Furthermore, we find observational evidence that the ULX might precess around its rotational axis, implied by a degree of degeneracy between different spectra observed at the same luminosity. Finally, we study the population of ULXs present in a sample of 17 nearby luminous infrared galaxies (LIRGs). It is found that the LIRGs possess significantly fewer ULXs per unit star formation rate than nearby normal galaxies, by a factor of about 10. We argue that part of the deficit could be due to the high metallicity environment of the host galaxies suppressing the formation of ULXs, and the lag between star formation starting and the appearance of ULXs; however, the majority of the deficit of ULXs is likely to be due to the high amount of gas and dust in the LIRGs obscuring a large fraction of ULXs.

One vital theoretical tool in the ever-improving description of particle physics is Monte Carlo event simulation. With the CERN LHC currently exploiting the highest energies in any human-made experiment, with unprecedented precision for a range of processes, very precise theoretical models are strongly motivated. This thesis introduces an implementation of Sudakov logarithms, which are a high-energy approximation to the exact NLO calculation of the electroweak quantum theory, within the Sherpa Monte Carlo framework. As well as this, it validates and applies the Sherpa + OpenLoops interface to a range of interesting electroweak processes at NLO in QCD. One key area of study in the electroweak sector is that of high multiplicities of weak bosons in the final state, which are motivated by the insight these studies could provide into the recently discovered 125 GeV particle, with properties consistent with the Standard Model Higgs boson. Furthermore, several electroweak processes are key backgrounds in searches for physics beyond the Standard Model. In addition, leading order results are presented for more complex electroweak processes for which such a high level of precision has not yet become necessary. The thesis is concluded with studies at a potential future 100 TeV proton-proton collider.

No description available.

No description available.

In this thesis we present analyses of z ~ 3 star-forming Lyman break galaxy and Lyman-alpha emitter populations. Additionally, we use QSO sightlines to probe the properties of gas around the LBGs. The observed star-forming galaxy sample is based on spectroscopic redshift data taken from a combination of the VLT LBG Redshift Survey (VLRS) data and Keck LRIS observations in fields centred on bright background QSOs. We compare these data with results from a GIMIC hydrodynamical simulation. We first estimate the auto-correlation function of simulated galaxies and compare these results with the observed Keck+VLRS correlation functions. We find that the observed galaxy real-space auto-correlation function is more consistent with that measured for simulated M⋆ ≥ 10^9 h^−1 M⊙ galaxies than lower mass galaxies. We then calculate the cross correlation of galaxies with Ly-alpha absorption in QSO sightlines in both our observed and simulated datasets. We check near star-forming galaxies in both data and simulations for the existence of the transmission spike which previous authors have claimed to be indicative of the effects of star-formation feedback on the IGM. No detection of such a spike is seen in the galaxy-gas correlation function in the combined VLRS+Keck data. The simulated cross-correlation function also shows comparable neutral gas densities around galaxies as seen in the observations. The Ly-alpha auto- and cross-correlation functions in the GIMIC simulations, appear to show infall smaller than implied by the predicted infall parameter of β_Lyα ≈ 1.3 (McDonald et al.). There is a possibility that the reduced infall may be due to the galaxy wide outflows implemented in the simulations. We present the Lyman-α luminosity functions and two-point clustering correlation functions of Lyα emitters (LAEs) at z = 3.1. We obtain a photometric sample of ~500 LAE candidates from 5 fields based on deep Subaru Suprime-Cam imaging data and a spectroscopic sample of 62 confirmed LAEs in 3 of our 5 fields from VLT VIMOS spectroscopy. We find that our narrowband Lyα luminosity function is in agreement with Ouchi et al. (2008) and is higher than for Gronwall et al. (2007). We estimate the R-band continuum luminosity functions of our sample of LAEs. Our 1700 Å continuum magnitude (auto) LAE luminosity function appears similar to that of Gronwall et al.(2007) with a relatively high number density of LAE being detected at bright magnitudes. We compare LAE and LBG R-band continuum luminosity functions and find that the LAE luminosity function at R < 25.5 amounts to only ~30 % of the density of the LBG luminosity function. Nevertheless, most of LAE still lie at the faint end of the LBG luminosity function. Finally, by comparing the 3-D LAE and LBG clustering amplitudes as estimated from the angular correlation function, we find a lower r_0 for LAE than for the LBGs. We measure a correlation length in the range of r_0 = 1.8-3.1 h^-1 Mpc and a bias in the range of b = 1.4-2.3 for LAEs. Even the highest LAE r_0 value is significantly lower than that of LBGs. The average LAE dark matter halo mass inferred from clustering is in the range of 10^{9.7±0.8}-10^{11.2±0.2} h^{-1} M⊙. The lower correlation length leads to the lower halo masses. Our results show that the LAE population is dominated by galaxies fainter than those traditionally selected via the Lyman Break method. It is possible that the LAE population may simply be a faint subset of the LBG selection and our results, in terms of both the luminosity function and clustering, are consistent with this picture.

This thesis reports the investigation of the scattering properties of a mixture of Rb and Cs and the formation of ultracold Feshbach molecules. The production of Feshbach molecules is a crucial step towards the production of ultracold polar molecules, which is of significant interest for a wide range of potential applications. We have investigated the scattering properties of a mixture of ⁸⁵Rb and ¹³³Cs in their lowest spin channel in a magnetic field range from 0 to 700 G. Furthermore, we explored the Feshbach spectrum of ⁸⁵Rb alone in both, the (f = 2,m_f = -2) and (2,+2) states up to a magnetic field of 1000 G. Additionally a Feshbach resonance in a (2,+2)+(3,+3) spin mixture was experimentally confirmed. We associated Cs₂ Feshbach molecules using a Feshbach resonance at 19.9 G. 2.1(1) x 10⁴ molecules with a temperature of ~ 60nK were produced from a sample of 3.28(2) x 10⁵ Cs atoms with a PSD of 0.20(1). Due to a magnetic field gradient, the molecules 'bounce' at an avoided crossing between two states at 13.5G. This gradient field was also used to produce multiple molecular clouds from one atomic sample. A combination of both techniques led to a 'collision' between two Cs₂ clouds. Furthermore, we associated up to ~ 5000 heteronuclear ⁸⁷RbCs Feshbach molecules using an interspecies resonance at 197.1G. Confined in the dipole trap the molecules have a lifetime of 0.21(1)s. We have measured the magnetic moment of the molecules in different internal states in a magnetic field range from 181 to 185G. Molecular loss spectroscopy on electronically excited states was performed to identify candidates for the intermediate state of a STIRAP transfer of the molecules into their rovibrational ground state. Subsequently, the binding energy of the rovibrational ground state was measured to be 3811.574(1)cm⁻¹, using two-photon spectroscopy.

The structural and photometric properties of ‘red-and-dead’ early-type galaxies provide vital clues about the evolutionary pathway which lead to their formation. Here, I use deep Canada-France-Hawaii Telescope image data to explore the multi-component internal structures of red sequence galaxies in the Coma cluster, with a particular focus on disk-dominated early-type galaxies (i.e. S0s). Galaxies are investigated across a wide range of uminosities (−17 > M_g > −22) and cluster-centric radii (0 < r_cluster < 1.3 r_200). I present the 2D structural decomposition of u, g, i imaging via GALFIT. Rigorous filtering is applied to ensure that the measured best-fit models are the most meaningful descriptions of their galaxy’s underlying stellar structures. A sample of Coma cluster members (N = 200) was identified as well described by an ‘archetypal’ S0 structure (central bulge + outer disk). Internal bulge and/or disk colour gradients were implemented by allowing component sizes to vary between bands. Such gradients are required for 30% of archetypal S0 galaxies. Bulges are characterised by n ~2 profiles with half-light radii, R_e ~1 kpc, remaining consistent in size for all but the brightest galaxies (M_g < −20.5). S0 disks are brighter (at fixed size, or smaller at fixed luminosity) than those of star-forming spirals. Similar colour-magnitude relations are found for both bulges and disks. The global red sequence for S0s in Coma hence results from a combination of both component trends. The average bulge − disk colour difference is 0.09 ± 0.01 mag in g − i, and 0.16 ± 0.01 mag in u − g. Using simple stellar population models, bulges are either ~2-3× older, or ~2× more metal-rich than disks. The trend towards bluer global S0 colours observed further from Coma’s core is driven by a significant correlation of disk colour with cluster-centric radius. An equivalent trend is detected in bulge colours at a marginal significance level. An environment-mediated mechanism of disk fading is favoured as the dominant factor in S0 formation. The decomposition analysis was then extended to encompass a wider range of structural models. This revealed a large sample of reliably-fit, symmetric multi-component galaxies in Coma (N = 478). 42±3% of Coma cluster galaxies (N = 201) are best described by a 3(+) component structure. In addition, 11% of galaxies (N = 52) feature a break in their outer profiles, indicating truncated or anti-truncated disks. Beyond the break radius, truncated disks are consistent in structure with untruncated disks, disfavouring a formation mechanism via physical truncation of exponential disks. The sizes/luminosities of bulges in antitruncated galaxies correlate strongly with galaxy luminosity, indicating a bulge-enhancement formation mechanism for anti-truncated disks. Both types of broken disk are found overwhelmingly (> 70%) in barred galaxies, suggesting that galaxy bar play an important role in formation of such structures. The wide variety of galaxy structures detected in Coma highlights the naivete of the simple ‘bulge + disk’ or ‘single spheroid’ morphological paradigm for early-type galaxies.

Developing a plasma exhaust solution suitable for future high power tokamaks is one of the major challenges facing the development of magnetic confinement fusion as a terrestrial energy source. In order to improve our understanding of the relevant physics, high quality experimental measurements of plasma dynamics in the scrape-off-layer (SOL) and divertor plasma regions are required. This thesis is concerned with the development of diagnostic instrumentation for measuring exhaust plasma flow: an important phenomenon with implications for the control of exhaust particles and heat as well as unwanted impurities. Coherence imaging spectroscopy (CIS) is a relatively new diagnostic technique which can be used to obtain time resolved 2D imaging of flows using the Doppler shifts of visible ion emission lines. The technique makes use of an imaging polarization interferometer and is based on the concept of Fourier transform spectroscopy. The principle advantages of this over other flow measurement techniques are the very large amount of spatial information collected, and the simple relationship between the measured quantities and spatially varying flows in the plasma. This thesis presents the development of, and first results from, a CIS ion flow diagnostic for the UK's Mega Amp Spherical Tokamak (MAST). The diagnostic can image flows of intrinsic C II, C III and He II impurity ions over fields of view between 10 - 45 degrees, at frame rates between 50Hz - 1kHz and with flow resolution typically around 1km/s (compared with measured flows of typically 5 - 30km/s). Spatial resolution is better than ~4.5 cm over a 1.4 x 1.4m area of the plasma cross-section. After reviewing the principles and theory of the coherence imaging technique, the design of a coherence imaging flow diagnostic for MAST is presented in detail. Results of careful laboratory characterization and calibration of the instrument are presented, and the instrument performance is compared to the design calculations. The diagnostic was used successfully for flow measurements on MAST during an experimental campaign in May - September 2013. On-plasma validation of the instrument performance is presented, as well as examples of novel flow observations made with the diagnostic. These include field-aligned flow structures associated with high field side gas fuelling of the plasma, and the first measurements of spatial flow structure in the divertor associated with the application of resonant magnetic perturbations (RMPs). Possible future improvements to the instrument design and extensions of the present work are suggested.

Astrophysical black holes should be simple objects with only two parameters: mass and spin. As material accretes onto the black hole this adds two further parameters: accretion rate and, since accretion generally occurs through a preferential plane, the inclination at which we view the system. Inclination becomes particularly important when a fraction of the inflowing material is not accreted but is instead ejected from the system in powerful, highly collimated and sometimes highly relativistic jets. It is these luminous accretion flows and jets that allow accreting black holes to be detected across the entire range of the electromagnetic spectrum from radio up to gamma-ray energies. The emission from the accretion flow and jet should be completely determined by the four fundamental parameters of mass, spin, accretion rate and inclination. Variations in these four parameters should be all that is required to explain the enormous variety of spectra from accreting black holes. In this thesis I present five papers studying emission from black holes of all size-scales and in all accretion regimes, including inclination effects and investigating the possible effect of the most difficult parameter to measure --- black hole spin. Black holes do not exist in isolation. Stellar mass black holes are fed by their companion stars and supermassive black holes by gas from their host galaxies. Not only does the galaxy fuel the growth of the supermassive black hole but equally the outflows that result from black hole accretion affect the growth of the galaxy, heating gas and suppressing star formation. This adds extra importance to understanding black hole accretion and the interplay between accretion and ejection. I find a scenario where low spin black holes are limited to feeding back via winds and moderately relativistic jets, while only the highest spin black holes are capable of producing the most powerful, highly relativistic jets, may be consistent with current observations.