Structure determination of thin polymer films using GIXRD and AFM Both atomic force microscopy (AFM) and grazing incidence x-ray diffraction (GIXRD) are used to determine the structure of polymer thin films. To enable experiments using these techniques to be conducted experimental apparatus was designed and built. These apparatus' included an in-situ heater to allow investigations at elevated temperatures to be conducted as well as a chamber for the work conducted using GIXRD to minimise the background scattering incident on the detector. The structure of two polymers, F8 and PQT-12, have been investigated using GIXRD. Both of these polymers align in a layered geometry normal to the surface, and in an isotropic polycrystalline ordering in the plane of the surface. The F8 polymer, while in the a-crystalline phase, has a unit cell dimension of 27.3 0.3A normal to the surface and unit cell dimensions of 24.3 0.3A and 17.1 0.2A in the plane of the surface. As- spun PQT-12 of molecular weight of 5000amu has been found to form a metastable self- assembled crystalline phase with unit cell dimensions of 35.6 0.7A out-of-plane and 13.10 0.12A, 16.4 0.6A in-plane. The structure of PQT-12 has been shown to be dependent on molecular weight, annealing and physical preparation of the substrate. The rate and mechanisms of crystallisation of a third polymer, PET, have also been investigated as a function of anneal temperature. The crystallisation was followed at each given temperature to follow the growth of individual spherulites. As the thickness of the PET film was increased from 43nm to 120 nm the radial growth rate of the spherulites increased linearly from 8.0 0.9nm/min to 14.3 1.3nm/min.

Electronic structure calculations are performed on metal surfaces using an embedding method. Firstly, Cu and Ni surfaces with adsorbates are studied. The O/Cu(00T) reconstruction is investigated, and it is found that atomic displacements increase bonding symmetry and is most likely the cause of reconstruction. The interaction between a single graphitic layer and the Ni(III) substrate is also studied, and it is found that interacting states are formed at Ni band gaps. The remainder of the thesis deals with steps on metal surfaces. First, the jellium model is used to calculate the work function dependence on step density. In the low step density limit, the work function varies linearly with step density. Further calculations are performed on realistic Pd and Pt surfaces. When electric fields are applied, the screening charge of stepped surfaces is mostly located at the step edge. This is explained by the increased fields associated with the edge. Field emission from Pd and Pt surfaces is also studied. It is found that Pd is a better emitter than Pt, owing to work function effects. Transmission is dependent on the surface parallel wavevector and decreases with increasing wavevector. The reduced work function also plays a role at stepped surfaces, although the major influence stems from the reduced local work function at the step site. The low effective potential at the step provides a locally reduced barrier to electron removal. In addition, an increase in transmission is seen from non-zero wavevectors for stepped surfaces. The result is that stepped Pd and Pt surfaces, with a step density of one step in every three (001) unit cells, show field emission increased by a factor of four compared with the flat (001) surfaces.

This thesis presents various studies into the effects of both growth and processing conditions on a) the electrical properties of interfaces of the CdTe solar cell, and b) the photovoltaic (PV) performance. Annealing of the CdS/TCO (transparent conductive oxide) bilayer in both oxidising and reducing ambients was investigated in order to study changes in the electrical properties of the In/CdS/TCO structure. It was found that post-growth oxidising changed the current–voltage (J-V) characteristics from Ohmic to rectifying, which was attributed to the creation of a CdO-n+/CdS-n junction, as an oxygen-rich layer was revealed by Auger electron spectroscopy (AES) on the CdS surface. A new method of testing pinholing of the CdS film was used, which gave confidence that the observed Ohmic behaviour was genuine. Annealing CdS (in various ambients) was further investigated by studying its effect on full devices, but the effect on PV performance was insignificant. This study was paired with an investigation into a rapid screening method of optimising CdTe/CdS cell PV performance, which reduced the number of the required samples by a factor of ~ 30. This was achieved by varying the CdTe thickness by chemically bevelling the cells in a Br2/methanol solution. The best performance was obtained at a CdTe thickness of ~ 3 μm, for which the CdCl2 treatment used was optimum. Both the uniformity and roughness of the cell layers are vital to obtaining high quality results using this methodology. The electrical current transport mechanism in the CdTe/CdS heterojunction was investigated as a function of a) growth technique by which devices were fabricated, and b) window layer type. Data were collected by recording J-V-T measurements in different light intensities (including dark), with temperature being varied in the range of 200 – 300 K. The transport mechanism was found to be dependent only on the window material under forward bias condition in the dark, but was independent of both the window layer and growth technique in a) forward bias in the light, and b) reverse bias in the dark. A new method was used to determine the diode ideality factor in the light, and therefore identify the transport mechanism. The back contact of the CdTe/CdS cell was investigated by measuring its barrier height (b). In a preliminary study, two methods of b measurement were compared, with the most applicable method being used to study Au, Sb2Te3 and As2Te3 contacts, with the CdTe back surface being either a) as-grown, b) nitric/phosphoric acid etched or c) plasma etched. The value of the barrier height for each contact and its impact on the cell performance are presented and discussed.

The applicability of radio frequency magnetron sputtering (RFMS) for the development of: a) transparent conducting oxides (TCOs) and b) fully sputtered CdTe/CdS solar cells is demonstrated. TCO materials - In2O3:Sn (ITO), SnO2:F (FTO), ZnO:Al (AZO) and ZnO:F (FZO) - were investigated with respect to key deposition parameters in an attempt to generate films with low resistivities and high transmittances. Minimum resistivity values of 1.2 x 10^-4 Ohm.cm and 4.7 x 10^-4 Ohm.cm were achieved for films of ITO and AZO respectively while maintaining transmittances of > 80%. Such films are viable for incorporation into CdTe based solar cells as front contact layers. A model for the dielectric permittivity for TCO materials is presented based on classical Lorentz and Drude models of bound and free electron behaviour, and a model of inter-band transitions that describes behaviour in the vicinity of a direct band-gap. The model is successfully applied to the tting of transmittance data for TCO films and used to extract opto-electronic properties. The results of a fully-sputtered CdTe prototype device structure are presented; a maximum conversion eciency of 12.5% is achieved. Further investigations, via XRD, into the effect of sputter pressure on CdTe films indicates that a 10 mTorr Ar pressure is best for optimising device effciency. J-V-T and C-V measurements show that at room temperature, current transport in the sputtered devices is dominated by Shockley-Read-Hall recombination and that the CdTe layer, under zero applied bias, is fully depleted with a carrier concentration of 4 x 10^14 cm^-3. Cross-sectional SEM and TEM show that both CdS and CdTe layers undergo significant recrystallisation during post-deposition CdCl2 treatment. A multi-layer optical model of transmittance is developed based on a transfer matrix method and using optical data acquired from spectrophotometry and ellipsometry. The model is used to predict the fraction of transmitted light received by the CdTe absorber in a fully sputtered device and it is indicated that through further thinning of the CdS layer, to below 50 nm, significant gains in transmittance, upwards of 20%, may be achieved. It is established that the further development of sputtered CdTe/CdS solar cells requires a significant improvement in the uniformity of the current CdCl2 based post-deposition treatment.

The behaviour of polymeric amphiphiles, specifically systems containing poly( methyl methacrylate), PMMA, and poly(ethylene oxide), PEO, spread at the air water interface is reported. An amphiphilic graft copolymer containing a PMMA backbone and PEO side chains forms the main subject of the study. Films of PMMA spread onto a 0.1 wt.% solution of PEO have also been considered. Due to the presence of a surface excess, both PEO and PMMA populate the surface and the latter system provides a model of the graft copolymer. Monolayer behaviour has been characterised by recording surface pressure isotherms. It has been demonstrated that the behaviour of the graft copolymers is highly composition dependent, exhibiting characteristics associated with the constituent homopolymers. In contrast the surface pressure isotherm for PMMA on PEO solution is quite unlike that for PMMA on water. Details of the structural organisation in these systems have been determined using neutron reflectometry. The surface organisation adopted by the graft copolymers varies as the surface concentration increases. The PEO side chains penetrate the subphase and stretch increasingly whereas the PMMA backbone behaves much like the homopolymer. This is in contrast to the behaviour of PMMA on PEO solution where the PMMA occupies a wider region of the interface. The structure of the adsorbed PEO differs from that of the bare solution but maintains a constant surface excess. Dynamic behaviour has been studied using surface quasi elastic light scattering. For the graft copolymers surface wave evolution is very unusual. The dilational parameters provide evidence, e.g. a negative dilational viscosity, indicating that surface wave propagation is complex and not described fully by current theories. Similar behaviour is exhibited by PMMA on a PEO solution. These features are believed to be associated with subphase penetration and are discussed in this context.

This thesis represents a systematic study of polycrystalline CdS thin films and their role as an n-type window layer in CdS/CdTe photovoltaic devices. This work encompasses the growth of CdS, primarily by the solution deposition method, and the subsequent characterisation of these films in isolation and as part of thin film CdS/CdTe device structures. A novel solution deposition approach was devised in order to grow high quality CdS thin films. Structural, electrical and chemical characterisation methods applied to these have shown that in their as-grown state they are highly oriented (in either the c[l1l] or h[002] direction), possess a small grain size of approximately 10-15 nm, and contain a considerable level of compressive strain. Annealing treatments in the presence of the fluxing agent CdCl(_2) have been shown to strongly modify the properties of these films, they are converted to a polycrystalline hexagonal structure with a significantly reduced level of strain, possess a larger grain size (27-28 nm) and a considerably enhanced crystalline quality. Novel 'hybrid' films comprising two CdS layers grown by different growth methods, one grown directly upon the other, have been studied. It has been shown that there are remarkable differences in morphology between these and films grown by a single growth method alone. Complete CdS/CdTe devices have been fabricated from several types of film grown in this study. Cell efficiencies of 9.80% were attained for a limited batch of devices, suggesting that these films possess good qualities for PV device fabrication. Early results from a novel tubular photovoltaic device concept are presented. This geometry has the potential to reduce manufacturing costs, may open up new routes to enhance the efficiency of CdS/CdTe devices, and is an attractive candidate for PV/solar thermal power generation.

Measurements have been made of the electrical resistivity in the film plane of rare earth metal films over a thickness range of 130 A to 250O A. The films are deposited onto glass substrates by high vacuum evaporation at pressures better than 10(^-6) torr and are protected by an over coating of silicon monoxide. The resistivity has been examined between 4.2 K and room temperature; the magnetoresistance has also been studied in the presence of applied transverse fields up to 15 KOe and longitudinal fields up to 10 KOe. The variation of resistance with temperature is similar in form to that obtained for bulk specimens for all film thicknesses. However, in films less than about 400 A thick, the spin-disorder resistivity shows a pronounced decrease in magnitude but the degree of variation did not appear to be consistent with thickness. The spin-disorder resistivity is obtained by computing a least-squares fit to the linear high temperature part of the resistance curve and extrapolating to 0 K. The results show that in the temperature region where spin-wave theory is applicable, at very low temperatures pspin ɚ T(^2) in accordance with spin-wave theory but at higher temperatures, this falls off, so that It is found that pspin ɚ T(^2) in the first case, the variation of Pspin(T) can be better explained in terms of an anisotropy energy gap and the change to a T(^3/2) proportionality has been related to the behaviour of the spontaneous magnetization and the anomalous Hall effect. In addition, the variation of pSpin of dysprosium is examined in detail and it is suggested that the observed fall off of spin resistivity with thickness, which can be extrapolated to zero at zero thickness, is related to the variation of magnetization with thickness in the film. Finally, magnetoresistive studies detected negative effects with fields applied transversely to the electric field. The anomalous peak at the Neel point is suppressed in dysprosium and terbium and a small demagnetizing field effect is observed in dysprosium.

In recent years it has become clear that knowledge of triplet transport in single layer and multilayer films can be crucial in improving the efficiency of organic light emitting devices and solar cells. This thesis reports an investigation of triplet exciton dynamics in small organic molecule single and multilayer layer films using optical time resolved nanosecond spectroscopy. A diligent step by step approach, leading towards the investigation of complex donor/spacer/acceptor multilayer structures is used. First of all, single layer films to be a constituents of multilayer structures were studied by measuring fluorescence, delayed fluorescence and phosphorescence. 4,4’-N,N’-dicarbazolyl-1,1’-biphenyl (CBP) widely used in organic light emitting diodes is characterized. Evidence is provided that in some of these spectra emission comes from trap states rather than the CBP molecule itself. Also N,N’-diphenyl-N,N’-bis(1-naphthyl)-1,1’-biphenyl-4,4’’-diamine (NPB) has been investigated. Results indicate that bimolecular triplet recombination is dominant and that triplet transport has dispersive features even at room temperature in NPB films. Then films of heavy metal iridium complexes fac-tris(2-phenylpyridine) iridium (Ir(ppy)3) and iridium(III)tris(1-phenylisoquinoline) (Ir(piq)3) are put into the spotlight. New states previously not reported are identified and decay with the slope -1 characteristic of more than one iridium complex and previously not published in literature is observed. Triplet interface sites in bilayer Ir(piq)3/NPB films obstructing triplet migration are determined and triplet movement across interface is experimentally captured for the first time. The origin of these interface trap states is suggested. Then this system is upgraded into Ir(ppy)3/NPB/Ir(piq)3 and triplet transfer from Ir(ppy)3 to Ir(piq)3 via NPB is investigated. A model of triplet exciton dynamics in Ir(piq)3/NPB films using classical diffusion equations is presented with interface sites included. Computer simulations were performed and the results are in very good agreement with the experimental ones. Finally problems encountered are identified and main guidelines on how to do research in complicated multilayer structures are set.

A scanning tunneling microscope has been used to study the surface properties of CdTe crystals. Coupled with scanning tunneling spectroscopy (STS) and atomic force microscopy, the surface morphology, structure and electronic properties of CdTe and CdZnTe surfaces have been studied. We have systematically investigated the three low index surface planes of the cubic crystal, that is the {100}, {110} and {111} surface planes. In addition, wet chemically treated surfaces were also examined. Clean surfaces were prepared in ultra-high vacuum conditions using argon ion sputtering and annealing. For each surface we imaged and recorded the surface reconstructions and morphologies. For the (100) surface, a mixed c(2x2)+(2x1) surface phase was found, where steps on the surface were found to preferentially align along <100> directions. For the (110) surface, tunneling spectroscopy was used to investigate the surface electronic structure of the (1x1) reconstruction. Using theoretical calculations of the tunneling current, we were able to match theory to experiment and discern the various vacuum tunneling processes for both n-type and semi-insulating material. For the (111) surface, a (2x2) reconstruction consisting of a cadmium vacancy structure was found. For the (-1-1-1) surface, a very disordered c(8x4) reconstruction was observed, consisting of a complicated tellurium terminated chain structure. For both faces, a large amount of faceting was observed to occur with the facets formed by {311} planes. Finally, wet chemically treated surfaces, important for the construction of many semiconductor devices, were investigated. Here the change in surface morphology for a variety of different common surface preparation methods was observed and, using STS, various surface electronic states were identified.

This work begins from depositing the same organic light emitting diode (OLED) structure on the glass substrate and the Polyethylene terephthalate (PET) substrate. The result suggests that the transparency of the substrates, the microcavity effect arising from different anode thicknesses and the surface roughness of the anode all come into play to lower the device efficiency of the PET device. The Young’s moduli of the Tris-(8-hydroxyquinoline) aluminum (Alq3) and N,N'-Bis(naphthalen-1-yl)-N,N'-bis(phenyl)benzidine (NPB) are measured using the nanoindentation technique. Although the substrate effect cannot be completely filtered out due to the thin film thickness and the softness of base (PET), with a Young’s modulus of around 40-100 GPa, the small molecule layer can still be considered to be applied in the flexible devices. The stress in the indium-tin-oxide (ITO) of the multi layer OLED device when being bent is then calculated using this elastic parameter. A buffer layer inserted between the ITO and PET substrate with an optimized Young’s modulus and film thickness is suggested so that, according to theoretical calculations, the stress in the ITO layer can be reduced by up to 60%. Finally the flexible OLED devices were fabricated and their electroluminescence (EL) was measured when the OLED layers were applied with a tensile strain and compressive strain. It is observed that the compressive strain caused less damage than the tensile strain during the bending. It is also observed that a 4 µm thick buffer layer does protect the OLED device in terms of EL brightness while bending with the radius of curvature R = 6 mm. Without the buffer layer, the brightness of the device drops by 40% compares to merely 10% with this buffer layer. It is finally concluded that with the proper mechanical design of the device structure, the small molecule based OLED devices using the ITO as anode can still be considered as one of the candidates of the flexible display or lighting devices.