Achieving fast charge carrier transport in disordered organic semiconductors is of great importance for the development of organic electronic devices. Disordered organic materials generally show low charge carrier mobilities due to their inherent energetic and configurational disorder, and the presence of chemical and physical defects. Efforts to improve mobility typically involve chemical design and materials processing to control macromolecular conformation and/or induce greater crystalline or liquid crystalline order. Whilst in many cases fruitful, these approaches have not always translated into higher bulk mobilities in devices. Addressing the adverse effect on mobility of specific types of disorder or specific defects has proven difficult due to problems distinguishing the many such features spectroscopically and controlling their formation in isolation. In the three experimental Chapters following, we attempt to make clear links between the charge carrier mobility and the presence of specific structural defects or sources of energetic or configurational disorder. In the first experimental study, we investigate hole transport in a family of polyfluorenes based on poly(9,9-dioctylfluorene) (PFO). By controlling the phase formation of the materials through processing and by virtue of their chemical design, we examine the effect on transport of distinct material phases. Remarkably, we are able to isolate the effect of the single chain conformation of PFO known as the beta-phase and show that when embedded in a glassy PFO matrix it acts as a strong hole trap, reducing the mobility of the bulk material by over two orders of magnitude. By fabricating a device with negligible beta-phase, we demonstrate the highest time-of-flight mobility in PFO to date, at over 3 10-2 cm2/Vs. This study provides the first clear and unambiguous example of the effect on transport of a distinct conformational defect in a conjugated polymer. We also demonstrate the adverse effect on mobility of crystallinity in the polyfluorenes. We suggest that our findings may generalise to other systems in the sense that the mobility may be limited by a minority population of structural traps, which may include highly ordered, crystalline regions. Significant mobility improvements may then be more easily achieved by removing the minority ordered phases than by increasing their concentration. We believe that this approach offers an alternative paradigm by which higher mobilities may be obtained in general, and in particular in systems where crystallinity is undesirable. In the second experimental study, we study charge transport in the fullerene derivatives [6,6]-phenyl-C61-butyric acid methyl ester (PCBM), bis-PCBM and tris-PCBM. The fullerene multi-adducts bis-PCBM and tris-PCBM are of interest as alternative OPV acceptor materials with the potential to increase open-circuit voltage. However, most OPV blends employing the multi-adducts have failed to improve upon those employing PCBM. This is thought to be a result of the inferior electron transport properties of the multi-adducts, due to either (i) higher energetic disorder in the multiadducts due to the presence of isomers with varying LUMO energies or (ii) higher con gurational disorder due to a lower degree of order in molecular packing in the multi-adducts than in PCBM. We distinguish the e ects of energetic and con gurational disorder using temperature-dependent ToF and FET measurements. We find that differences in configurational disorder appear negligible, and that the reduced mobility in the multi-adducts is due predominantly to the energetic disorder resulting from the presence of a mixture of isomers with varying LUMO energies. In the third and final experimental study, we examine the charge transport properties of polymer: PCBM blends for OPV, focusing on the PTB7:PCBM and P3HT:PCBM systems. In particular, we address the question of why state-of-the-art OPV systems such as PTB7:PCBM perform so much worse at large active layer thicknesses than P3HT:PCBM. We find that low electron mobility is the main cause of this di erence. The electron mobility in PTB7:PCBM blends, at 10-5 { 10-4 cm2/Vs, is 1-2 orders of magnitude lower than the electron mobility in annealed P3HT:PCBM, at over 10-3 cm2/Vs. The hole mobility, in contrast, is the same to within a factor of approximately three. We hypothesise that the low tendency of PTB7 to order leads to a low degree of phase separation in the blend and to a poorly connected, disordered PCBM phase. We find that increasing the PCBM fraction is very effective in improving electron transport and electrical Fill Factor, but strongly reduces absorption. We suggest that a key challenge for OPV researchers is thus to achieve better connectivity and ordering in the fullerene phase in blends without relying on either (i) a large excess of fullerene or (ii) strong crystallisation of the polymer.

This thesis summarizes the results and challenges of a series of experiments in the field of atomic and molecular physics carried out employing innovative light sources such as X-ray free electron lasers, high harmonic generation sources and ultrafast mid-IR lasers. The key feature common to all of them is the ability to provide ultrashort light pulses with a pulse duration in the 1-100 fs range (1 fs = 10-15 s) which in principle allows the investigation atomic and molecular dynamics unfolding on such time scales. The experiments described in this thesis constitute the first steps in this direction and shed light on the new challenges and opportunities that arise naturally when highly innovative tools are employed. In the main technological chapters, which follow a brief description of the various light sources, a variety of experimental techniques will be described, such as velocity map imaging, electron Time-of-Flight spectroscopy, ion Time-of-Flight mass spectroscopy and covariance mapping. Among these velocity map imaging is the one which has been used more extensively. A VMI spectrometer for electrons with kinetic energies in the 0-200 eV range has been designed and tested for the Artemis Lab at Rutherford Appleton Laboratories. For the analysis of the output images two well established algorithms for Abel inversion have been implemented and compared. In the first experimental chapter one of the first applications of the new Artemis VMI spectrometer will be described. In the TRPEI (time-resolved photoelectron imaging) Artemis experimental campaign the radiation produced by the monochromatized HHG beamline was employed to study the photoelectron angular distribution of electrons ejected from valence and inner valence shells in a number of atomic and molecular samples. Each of the remaining experimental chapters will be devoted to an experiment carried out at the first Hard X-ray Free Electron Laser facility in the world: the Linac Coherent Light Source. In the chapter devoted to the ring opening of 1,3-cyclohexadiene (CHD) a complex molecular reaction, namely the conrotatory electrocyclic opening of CHD to form the linear isomer 1,3,5-hexatriene, is studied on a time scale of 1 ps by X-ray induced fragmentation. Double core hole creation is the subject of the following chapter where the process is investigated with covariance mapping. Finally X-ray induced molecular dynamic following core excitation of molecular oxygen is addressed in the last chapter.

This thesis aims at analysing the impact of oceanic turbulence and air-sea interactions on the sea surface temperature (SST) of the extra-tropical oceans on spatial scales of a few hundred kilometres (the so-called "mesoscale"). Using satellite-based measurements of SST and sea level, as well as surface tracks of mesoscale oceanic cyclones and anticyclones, it is shown that turbulence does not transport heat through systematic motions of cold cyclones and warm anticyclones, as was previously thought in regions of strong mean flows like the Gulf Stream. Rather, it is suggested that heat is transported as a result of the slight phase shift between temperature and pressure fluctuations developing on the mean flow. In addition, tentative estimates of the rate at which air-sea heat fluxes damp the SST signatures of cyclones and anticyclones are provided. The weak values obtained ( 20 W/m2K) contrast sharply with theoretical expectations, but are in agreement with the observed long-lived thermal heat content anomalies associated with the cyclones and anticyclones. These observations provide important benchmarks for high-resolution ocean models and may moreover guide the parametrization of subgrid-scale heat transport in climate models.

This thesis reports on the development of an array of plane-concave Fabry-Perot microcavities containing atoms (or other quantum emitters), interconnected by UV-written waveguides on a silica-on-silicon chip. The microcavities are formed by a mirror coated on the end facet of the chip and an array of spherical micromirrors etched on silicon. This is to our knowledge the first attempt at implementing the emerging coupled-cavities QED paradigm. The device we propose possesses a degree of control, flexibility and tuning unmatched in other suggested implementations: The atoms can be manipulated inside the cavity by auxiliary lasers and the cavity-cavity coupling rate as well as the atom-cavity coupling can be tuned. It is highly scalable. Calculation of the complete (classical) optical spectrum of the device is presented. The quantum dynamics that may eventually be observed has also been studied. Waveguide chips containing couplers and phase shifter have been fabricated. We have successfully demonstrated the operation of the elementary sub-systems: the strong optical coupling between a microcavity and a waveguide resonator, and the tunable strong coupling between two evanescently coupled waveguide resonators. No experiments with atoms or other quantum emitters were attempted, because the waveguide propagation loss is so large that no quantum physics can be observed. There is hope that this can be overcome in the future by using other waveguide technologies.

The UKMO has tested the effect of assimilation of aerosol optical depth (AOD) observations from the SEVIRI retrieval algorithm which has been developed at Imperial College and this has been shown to add skill to its short term forecasting ability. This thesis details the validation of the SEVIRI AOD retrieval algorithm prior to a longer assimilation trial and considers both internal and external factors which may affect its success using data from the GERBILS airborne field campaign during June 2007. It was found that of the factors tested, the retrieved AODs were most sensitive to the coarse mode in the size distribution assumed in the radiative transfer modelling used to derive the expected dust signal. Following on from this work, co-located SEVIRI retrieved AODs and GERB fluxes at the top of atmosphere (TOA) were used to provide, for the first time, an observationally based estimate of the net direct radiative effect (DRE) of mineral dust over North Africa from geostationary satellite observations, giving new insight into the influence of time of day on the SW, LW and net effects. As expected, mineral dust aerosol was found to reduce the outgoing longwave radiation at all times of day with the peak reduction clearly following the diurnal cycle of surface temperature. Instantaneous SW DRE was found to be large under certain conditions and showed strong dependencies on pristine sky albedo and solar zenith angle such that a given aerosol loading could induce a positive or negative SW DRE. Overall, the mean SW DRE over June 2007 was found to be negligible. The net DRE for June 2007 was hence found to be dominated by the LW component with mineral dust inducing a reduction in outgoing net flux of the order 10Wm-2. Besides being of interest in their own right, the methodology and results from this analysis should help to provide an important constraint on the success of the assimilation process in terms of assessing whether the assimilated aerosol has the 'correct' impact on the forecast radiative fields.

Cosmological inflation is the dominating paradigm to account for observations of the Cosmic Microwave Background (CMB). In this thesis, we study the phenomenology of a class of particularly well motivated models of inflation, known under the generic name of hybrid models. They are characterised by a transition from a valley to a hilltop shaped potential. In particular, we study three limiting regimes of the simplest realisation, hybrid inflation, constraining its parameter space using observational bounds on the spectral index and the non-gaussianity of the primordial perturbations. We find that the model is highly constrained by observations, with large part of the parameter space either ruled out by a blue spectral index (ns > 1) or by a large non-gaussianity parameter fNL, two quantities measured with precision by PLANCK. However, there exists regions in parameter space leading to interesting phenomenology compatibly with observational bounds. Also, a version of hybrid inflation with a third light scalar field at horizon crossing is derived from the supersymmetry framework. We find that the model can generate observables within observational bounds.

This work presents recent developments on brane tilings and their vacuum moduli spaces. Brane tilings are bipartite periodic graphs on the torus and represent 4d N = 1 supersymmetric worldvolume theories living on D3-branes probing Calabi-Yau 3-fold singularities. The graph and combinatorial properties of brane tilings make the set of supersymmetric quiver theories represented by them one of the largest and richest known so far. The aim of this work is to give a concise pedagogical introduction to brane tilings and a summary on recent exciting advancement on their classification, dualities and construction. At first, particular focus is given on counting distinct Abelian orbifolds of the form C3/[gamma]. The presented counting of Abelian orbifolds of C3 and in more general of CD gives a first insight on the rich combinatorial nature of brane tilings. Following the classification theme, the work proceeds with the identification of all brane tilings whose mesonic moduli spaces as toric Calabi-Yau 3-folds are represented by reflexive polygons. There are 16 of these special convex lattice polygons. It is shown that 30 brane tilings are associated with them. Some of these brane tilings are related by a correspondence known as toric duality. The classification of brane tilings with reflexive toric diagrams led to the discovery of a new correspondence between brane tilings which we call specular duality. The new correspondence identifies brane tilings with the same master space - the combined mesonic and baryonic moduli space. As a by-product, the new correspondence paves the way for constructing brane tilings which are not confined to the torus but are on Riemann surfaces with arbitrary genus. We give the first classification of genus 2 brane tilings, illustrate the corresponding supersymmetric quiver theories and analyse their vacuum moduli spaces.

The T2K experiment is a long baseline neutrino oscillation experiment. The most intense accelerator muon neutrino beam ever built is produced at the J-PARC facility in Tokai, located on the east coast of Japan, and directed (2.5 [degrees] off -axis) toward the Super-Kamiokande water Cherenkov detector, located 295 km away in the mountains of western Japan. For the current T2K [Symbol appears here. To view, please open pdf attachment]e appearance analysis, 23% of the background at Super- Kamiokande is from neutral current interactions that produce one π0 in the final state (NC 1π0). This study uses ND280 Run 2 data totalling 7.832 x 10 [to the power of 19] POT and selects neutrino-induced NC 1π0 interactions in FGD1 and FGD2 where both π0 decay photons convert in the barrel and downstream ECAL. Overall, neglecting photon reconstruction efficiency, an efficiency of 22% and purity of 16% is obtained. For the case where both π0 decay photons convert in the downstream (barrel) ECAL an efficiency of 33% (12%) and purity of 43% (9%) is achieved. When one photon converts in the barrel ECAL and the other in the downstream ECAL the efficiency (purity) is 29% (16%). Systematic uncertainties arising from cross-section model parameters and final state interactions, the neutrino flux prediction and the shape discrepancy between ECAL variables in data and MC are evaluated. The predicted total number of events when both photons convert in the downstream and barrel ECAL is 7.1± 2.3 (syst)± 0.5 (stat) and 31.6± 8.9 (syst) ±1.2 (stat), respectively. Whereas when one decay photon is reconstructed in the barrel ECAL and the other in the downstream ECAL the expected number of events is 38.1± 8.7 (syst) ±1.3 (stat). The selected number of events in data is 5, 22 and 31, respectively, which agrees with the prediction within errors. The calibration of the time offsets for the barrel and downstream ECAL is also presented in this thesis. The application of this calibration reduces the timing resolution from 9 ns to approximately 2.5 ns, for high charge hits.

Cosmic microwave background (CMB) radiation fills the universe almost uniformly and it has imprints of the primordial curvature perturbations. Observations of the CMB showed that to a large extent they are Gaussian [1]. However, there are hints of some non-Gaussianity in the CMB that is within the reach of experiments such as the Planck satellite. Different models of inflation predict different levels of non-Gaussianity. Comparison with the experiments could be used to falsify some models and single out the one describing the physics of inflation most correctly [2]. It has been found that preheating can produce substantial levels of non- Gaussianity in the massless model in the FRW backgroud [28]. In my thesis I present a detailed description of a refined calculation, which does not assume an FRW Universe but allows for the subhorizon inhomogeneous metric perturbations to interact with the scalar fields in a fully non linear fashion. Including the subhorizon metric perturbations in the dynamics is a physically much more accurate calculation. It allows for increasing the lattice size at will and therefore allows for including in the dynamics even those modes whose wavelengths are equal to or greater than the size of the horizon. I combined the lattice simulations with the separate universe approximation [5, 40]. The latter states that points in space separated by more than one Hubble distance cannot interact with each other and therefore evolve independently. The numerical integration of the Einstein Equations uses the BSSN scheme [7], a technique used to solve initial value problems in the 3+1 formulation of numerical relativity. I found that it is indeed possible to compute the superhorizon curvature perturbations at a constant density hypersurface including full metric perturbations. However, I also identified that accuracy improvements are needed to compare my results to the literature. I identified and successfully implemented a number of the improvements and presented some suggestions for the future research.

The unconventional long range (m=1) spin triplet proximity effect is theorised to be created at a Superconductor/Ferromagnet (SC/FM) interface that is magnetically inhomogeneous and provides the conditions for spin mixing. Many experimental and theoretical works have identified the existence of the m=1 spin triplet state using Josephson junctions to show how the critical current does not diminish significantly in a thick FM layer, (thickness being relative to the decay length of a Cooper pair in an FM). In this thesis the challenge was to examine single SC/FM, interfaces via point-contact Andreev Reflection spectroscopy (PCARS), to see if it were possible to find clear signatures of the spin mixing or spin scattering conditions that facilitates formation of the spin triplet state. PCARS measurements were performed on Ho single crystals and thin films as a function of contact resistance. The results show an anomalous decrease of the polarisation parameter and an increase in interface scattering parameter as the contact resistance decreases. This is interpreted in terms of the tip experiencing a region in which some form of spin mixing or spin scattering dominates. Measurements of Ni0.19Pd0.81 spectra were also taken as a function of contact resistance but the change of polarisation and scattering parameter that was observed in the Ho samples was not observed here. PCARS temperature measurements were also taken for Ni0.03Pd0.97 to investigate if spin fluctuations close to TC could provide the magnetic inhomogeneity required for spin-mixing. However the trend in the results proved inconclusive. This is likely due to broadening that result in the fitting procedure, and the extremely small polarisation in these samples. During the course of the thesis I contributed to the development of the spectral fitting routines and examined the challenges associated with three and four parameter fitting, degenerate fits and series resistance.