New generations of photovoltaics (PV) have demonstrated a significant cost-reduction with respect to c-Si wafer-based modules. Though second (thin-film) and third generation PV (high-intensity, low-cost) are already in the PV market, the preparation of standard procedures for their characterization is still ongoing. This work was developed by the author in order to extend some of the existing characterization techniques to a set of three different emerging technologies: multi-junction thin-film modules, concentrator PV cells and luminescent solar concentrators. An original method for the spectral response measurement of large area thin-film multijunction modules is presented in the first part: the method is validated with several examples. A basic theoretical approach is also presented to propose innovative explanations of measurement artefacts that are observed in the literature. In the second part of the thesis, the setup, characterization and classification of a high intensity pulsed solar simulator for concentrator PV cells is illustrated. A new procedure for the preparation of a set of filtered reference cells for the irradiance detection at high intensities is also presented, providing an original tool for the verification of the linearity of these devices towards irradiance, which is usually assumed in the literature. In the third part the performance characterization of high-efficiency luminescent solar concentrators is presented: a simple ray-tracing model and its experimental validation, the impact of backside diffusive reflector on the performance of this kind of devices are mainly highlighted. The work was developed in support of the activities of the European Solar Test Installation laboratory of the European Commission, a centre of reference for PV testing.

In this thesis, new mechanically robust, high performance transparent conducting films of commercially sourced arc-made Single-Walled Carbon Nanotubes (SWCNTs) on both glass and flexible substrates were produced using spin-coating or spray deposition, interlayer or stencil patterning methods and used for fabricating efficient, flexible polymer-fullerene bulk hetero-junction solar cells. After carefully optimizing the dispersion process of SWCNTs with H2O:SDS (up to 0.03 wt.%) and developing and efficient surfactant removal/p-doping procedure with nitric acid, highly conductive and smooth SWCNT thin films (ca. 30 nm) were obtained with more than 6,500 Scm-1 at > 69 % transmittance and 7 nm (r.m.s.) roughness. In particular, SWCNT films spray coated from H2O:SDS exhibited electrical conductivities of up to 7694 ± 800 Scm-1. To our knowledge, these values are the highest so far reported for SWCNT electrodes. Peak values for the ratio of the dc conductivity to the optical conductivity (σdc/σop) were obtained as up to 24, which is quite similar to state of the art SWCNT films so far reported. In addition, two patterning methods were developed to define electrode patterns of SWCNT thin films for electronic device applications. Interlayer lithography provided a fast and high resolution patterning procedure for SWCNT thin films at micron and sub-micron length scales, which is important for the fabrication of high-speed transistors requiring short channel lengths, and offers an attractive route to fabricating high-density integrated circuits. In addition, stencil patterning provides a simple and fast method, which is well suited for low resolution electronic device applications such as organic solar cells. The patterned highly conductive SWCNT electrodes were incorporated into P3HT:PCBM bulk heterojunction solar cell applications, obtaining the best device performance of 3.6 %, which is the best result so far reported in the literature. Finally, to break through the limited performance (σdc/σop < 25) of SWCNT thin films, layered hybrid thin films of SWCNTs on reduced Graphene-Oxide were fabricated by a simple spray coating method and the optimised hybrid films were incorporated into relatively efficient organic solar cells (2 % efficiency).

The measured photocurrent of a solar cell may be considered the sum of a photogenerated current due solely to the influx of photons, and a photovoltage-induced current due to carrier injection at the electrodes. Correcting the measured photocurrent for the injected current yields the voltage dependence of the photogenerated current alone. This corrected photocurrent can provide valuable insight into the processes governing the behaviour of a solar cell, yet is seldom measured or discussed within the community. In this dissertation, an original experimental technique designed specifically for the reliable measurement of the corrected photocurrent is described, with the intent of applying it to organic bulk-heterojunction solar cells. Solar cells based on a number of donor-acceptor combinations were investigated. Using the experimental technique developed here, corrected photocurrent-voltage characteristics exhibiting remarkably anti-symmetric profiles were obtained and subsequently rationalised with a simple physical model. From the perspective of this model, the nature of charge extraction at the electrodes - and how this is affected by processes such as thermal annealing - was examined. Finally, a new low band-gap, small-molecule acceptor material was used in bulk-heterojunction solar cells, and shown to promising photovoltaic performance. Interestingly, these devices exhibited anomalous current-voltage characteristics, which, on closer examination, could be explained by an electric field dependence in the photogeneration rate. Throughout this work, particular attention was given to how these findings may be used to improve device efficiencies.

Ever-increasing energy consumption and consequent extensive greenhouse gas (GHG) emissions are two major urgent problems faced by all human beings in the 21st century. As a major contributor, the energy production section appears to be the most suitable field where further improvements could be explored to tackle these problems. Polygeneration is a typical type of next generation energy production technology with higher energy efficiency and lower/zero GHG emissions. However, methodologies guiding an efficient and stable transition from our existing energy systems to more advanced ones are still lacking. The purpose of this thesis is to provide a generic modelling and optimization framework to guide planning and design of energy systems. This framework of methodologies ad- dresses the following issues arising in the planning and designing of energy systems: a) decision making at both strategic planning level and process design level; b) selection of roadmaps, technologies, and types of equipment from many available options; c) planning or design according to both economic and environmental criteria; d) planning or design under inevitable and unpredictable future uncertainty. The thesis is organized as follows: first, a review of energy systems is presented, followed by methodologies of energy systems engineering and their applications. Then a section of polygeneration process modelling is provided, at both strategic planning and process design levels, comprising superstructure representations of polygeneration energy systems at different levels, implementations of the superstructure based modelling strategy using mixed-integer programming, multi-objective optimization for the optimal process design according to both economic and environmental criteria, and optimization under uncer- tainty to account the impacts of future uncertainties at the planning/design stage and to increase the flexibility and robustness of a process design. Finally, major achievements of this work are summarised and future research directions are recommended.

Many new and renewable forms of electricity generation are small scale and geographically constrained by the resource they use. They are connected into the local electricity distribution network rather than the national transmission network and are known as Distributed Generators (DGs). The developers (or owners) of such DG choose the location and capacity of their plant to maximise the economic benefit that arises. However, the distribution network was not planned and designed to accommodate such DG and various problems can arise such as voltage magnitude disturbance, excessive power flow in certain lines, excess fault levels, reverse power flows through the Grid Supply Point (GSP) transformers and increased power losses. For this reason, the Distribution Network Operator (DNO) may limit the capacity of plant built or constrain its output under certain operating conditions. Such constraints clearly affect the economic case for the plant. Traditionally, the economic and technical aspects of DG plant planning have been carried out sequentially and not in an integrated fashion. This thesis investigates how to combine the two sets of analysis such that both sets of influences are brought to bear in one process in choosing an optimum plant capacity. By using the proposed methodology, the interests of both the DG owner and the DNO are served. The proposed Techno-Economic assessment tool has been developed for an example case of a biomass fuelled generator. The key factors considered in the economic analysis are biomass yield density, transportation costs, capital costs of plant and the value of unit electricity. An economically optimum plant can be found based on the optimum radius of collection area. The issue of network constraints have been investigated by using load flow analysis techniques with various case study networks. The networks are real examples from the UK distribution network, chosen to give a variety of meshed and radial structures and load densities. From the load flow analysis,indications of breaches of network constraints are generated and sensitivity indices are produced which allow the proposed DG plants within the area to be constrained in various ways depending on the optimisation objectives chosen. Conclusions are also drawn on the extent to which the network structure and the geographic arrangement of load in a network affect the siting and sizing of the optimal DG plant. Further factors that affect the economics of a biomass plant are also considered. These include an analysis of the effect of the physical shape as well as the location of the collection area. Other potential sources of revenue in addition to the sale of electrical energy, include incentives for renewable energy including carbon trading and Renewable Obligation Certificates (ROCs) and the potential to provide various ancillary services to the network. The effect that these additional sources of revenue may have on the Techno-Economic feasibility analysis have also been investigated.

Solid oxide fuel cells (SOFCs) are of increasing interest as low emission, high efficiency, energy conversion devices for the production of electricity, and in some cases heat, from a wide range of fuels. In general, a porous cermet of nickel/ytrria-stabilized-zirconia (Ni/YSZ) is used as an anode with a dense electrolyte of ytrria stabilized zirconia (YSZ), enabling operation at temperatures above around 750 oC. Nevertheless, operating at high temperature leads to various problems such as metal corrosion, electrode sintering, and unwanted interfacial diffusion in the cell. In this regard, cermet anodes such as nickel/samarium-doped-ceria (Ni/SDC) and nickel/gadolinium-doped-ceria (Ni/CGO) have been proposed as alternative anodes to be operated with SDC and CGO electrolytes at intermediate temperatures (< 700 oC), respectively. However, less attention has been given to nickel/scandia-stabilized-zirconia (Ni/ScSZ) cermet anodes as alternatives for intermediate temperature SOFCs, despite the fact that the ScSZ electrolyte exhibits a higher ionic conductivity compared to other zirconia electrolytes, which may offer some advantages, especially at lower temperature. Furthermore, Ni/ScSZ anodes have shown improved tolerance towards carbon deposition and sulphur poisoning in addition to improved durability when compared to Ni/YSZ anodes. To date there have only been limited studies into the relationship between the materials used, the processing conditions, and the properties and performance of Ni/ScSZ anodes. This thesis is therefore aimed at (i) studying the optimum fabrication conditions and properties of Ni/ScSZ anodes, (ii) investigating the effect of ingredients such as binder content, solvent type and solid content in the ink formulation on the rheological properties of NiO/ScSZ inks and their applicability for screen-printing and (iii) relating the rheological properties of NiO/ScSZ inks to the performance and properties of the resultant anode films. A large part of the work is focussed on the fabrication and rheological properties of NiO/ScSZ screen-printing inks. The properties are linked to the particle network strength within the formulated inks and relate to the microstructure, mechanical strength, electrical performance and electrochemical performance of the resultant anode films. Overall, Ni/ScSZ anodes having 40 vol% Ni were found to be optimum in terms of both electronic conductivity and electrode polarization resistance. The anode exhibited improved tolerance towards carbon deposition compared to Ni/YSZ at intermediate temperature (700 oC). The effects of binder and solid content on the rheological properties of NiO/ScSZ screen-printing inks were studied by evaluating the thixotropic properties, yield stress and viscoelastic properties of the inks. The study indicated improved thixotropic properties, yield stress and particle network strength within the inks as the binder and solid content increased. These improved properties can be related to better particle bridging within the inks. A percentage ink recovery of 40 to 65 % was determined sufficient for the production of quality films with minimum defects. From the study, inks having 26 vol% solid with 3 wt% binder or 28-30 vol% solid with 2 wt% binder were determined as optimum for screen-printing using squeegee load, squeegee length, printing speed, snap-off and screen type of 6 kg, 5 in, 0.02 m/s, 2 mm and 325, respectively. Furthermore, films fabricated using these inks revealed improved particle connectivity, higher electronic conductivity, lower electrode polarization resistance and improved mechanical hardness. The solvent type was determined to have only a small impact compared to the binder and solid contents in the inks.

In the search for solid oxide fuel cell (SOFC) materials, which exhibit improved oxide ion conductivity at lower temperatures, attention has recently been focussed towards materials with atypical structural chemistry and diffusion pathways. For example high oxygen diffusivity is reported in fergusonite structured CeNbO4+δ, which can incorporate a range of oxygen excess stoichiometries by oxidation of Ce3+ to Ce4+ and as a result, four commensurate or incommensurately modulated superstructures of the monoclinic parent cell are possible. An alternative strategy to reduce migration energy barriers is to change the charge carrier from oxide ions to protons, and through the introduction of oxide ion vacancies, develop materials that can conduct both species through a O2(g)/H+(g) partial pressure dependence. Alkaline earth doped rare earth niobates RE1-xAxNbO4-δ (RE=La, Nd, Gd, Tb, Er A=Ca, Sr, Ba) have been shown to exhibit proton conductivities that can be maintained to much higher temperatures compared to traditional perovskite materials, with reported stability to both protonic and acidic media. However they are hindered by low dopant solubility, and an intermediate temperature monoclinic to tetragonal phase transition that is problematic for device applications. Owing to the unique structural and redox behaviours of the cerium analogue of the RENbO4 series, this worked aimed to investigate the transport properties of acceptor doped Ca1-xAxNbO4±δ. This work has shown that Ce1-xAxNbO4±δ can cycle between oxygen hyper- and hypostoichiometry, and in the process alternate between mixed electronic interstitial oxide ion conduction, and (mixed electronic) protonic conduction respectively. Under both oxidising and reducing conditions the conductivity increases by over one order of magnitude relative to both CeNbO4+δ and the most protonically conductive of the series La1-xAxNbO4-δ respectively. This is a result of the greater solid solubility of strontium and calcium in CeNbO4+δ, relative to LaNbO4 (1-2%), and is speculated to arise from the charge compensation mechanism of electron hole formation via Ce3+-Ce4+ oxidation, where the change in both cerium valance and coordination may act to stabilise the dopant. The likelihood of electronic charge carriers in the hypostoichiometric phases of Ca1-xAxNbO4±δ, is of interest due to limited number of mixed proton-electron conductors if the properties can be optimised further. Furthermore the transition temperature raises from ~520 °C to ~650 °C on exchange of lanthanum with cerium. Unusually the hyperstoichiometric phases are also suggested to show proton conductivity, indicating Ca1-xAxNbO4±δ may exhibit more unusual defect reactions.

Chalcopyrite copper indium gallium diselenide (Cu(In,Ga)Se2 or CIGS) solar cells have achieved the highest laboratory power conversion efficiency among thin film PV technologies, and have the potential to be manufactured cost-effectively at large scale. In this thesis, detailed studies on the growth of CIGS solar cells by means of a pilot scale inline co-evaporation system have been performed to explore the beneficial deposition techniques to successfully transfer the research achievements into cost-effective industrial production. The work involved the development and optimisation of several new processes along with the design of the inline pilot scale system. Effects of the thin absorber layers (with thickness of ~1 μm) on material quality and the device performance have been investigated using (i) a two-step process with room temperature evaporated metal precursors, (ii) a three-stage process at constant substrate temperature and (iii) a three-stage process at varied substrate temperatures. Changing the copper and gallium compositions of the CIGS layers was observed to have a strong impact on the structural and electronic properties and hence on the performance of the solar cells. In one study, simple mechanical compression was employed on a batch of low quality porous CIGS films to significantly improve the surface morphological and the optoelectronic properties. Photoluminescence measurements revealed the band-to-tail defect-related recombination that was detrimental to the quality of as-grown and compressed films. In another study, rapid thermal processing (RTP) was applied to the low temperature Cu-In-Ga-Se precursor layers in an attempt to reduce the cost and optimise the reaction mechanism of the chalcopyrite material. After fine tuning of the process conditions, the CIGS layers with larger grain size exhibited power conversion efficiencies up to 10.8%. While these results are promising, the device performance was mainly limited by the low open circuit voltage and the low fill factor. It has been found that the Cu-In-Ga-Se precursors doped with low concentrations of sodium were beneficial for re-crystallisation of the CIGS films due to the reduced grain size and crystallinity of the precursors. An in-situ X-ray diffraction technique was used to investigate the phase evaluation in both Na-free and Na-doped Cu-In-Ga-Se precursor layers as a function of temperature. The results confirmed that the formation of both CIGS as well as secondary phases in the layer with the Na-doped precursors started at higher temperatures compared to Na-free precursors. It is therefore expected that further improvements in the solar cell efficiency might be achieved following this RTP and low-temperature precursor process.

Emerging smart-grid-enabling technologies will allow an unprecedented degree of observability and control at all levels in a power system. Combined with flexible demand devices (e.g. electric vehicles or various household appliances), increased distributed generation, and the potential development of small scale distributed storage, they could allow procuring energy at minimum cost and environmental impact. That however presupposes real-time coordination of demand of individual households and industries down at the distribution level, with generation and renewables at the transmission level. In turn this implies the need to solve energy management problems of a much larger scale compared to the one we currently solve today. This of course raises significant computational and communications challenges. The need for an answer to these problems is reflected in today’s power systems literature where a significant number of papers cover subjects such as generation and/or demand management at both transmission and/or distribution, electric vehicle charging, voltage control devices setting, etc. The methods used are centralized or decentralized, handling continuous and/or discrete controls, approximate or exact, and incorporate a wide range of problem formulations. All these papers tackle aspects of the same problem, i.e. the close to real-time determination of operating set-points for all controllable devices available in a power system. Yet, a consensus regarding the associated formulation and time-scale of application has not been reached. Of course, given the large scale of the problem, decentralization is unavoidably part of the solution. In this work we explore the existing and developing trends in energy management and place them into perspective through a complete framework that allows optimizing energy usage at all levels in a power system.

The development of flexible lightweight OLED devices requires oxygen/moisture barrier layer thin films with water vapour transmission rates (WVTR) of < 10-6 g/m2/day. This thesis reports on single and multilayer architecture barrier layers (mostly based on SiO2, Al2O3 and TiO2) deposited onto glass, Si and polymeric substrates using remote plasma sputtering. The reactive sputtering depositions were performed on Plasma Quest S500 based sputter systems and the morphology, nanostructure and composition of the coatings have been examined using SEM, EDX, STEM, XPS, XRD and AFM. The WVTR has been determined using industry standard techniques (e.g. MOCON) but, for rapid screening of the deposited layers, an in-house permeation test was also developed. SEM, XRD and STEM results showed that the coatings exhibited a dense, amorphous structure with no evidence of columnar growth. However, all of the single and multilayer coatings exhibited relatively poor WVTRs of > 1 x 10-1 g/m2/day at 38 °C and 85 % RH. Further characterisation indicated that the barrier films were failing due to the presence of substrate asperities and airborne particulates. Different mechanisms were investigated in an attempt to reduce the density of film defects including incorporation of a getter layer, modification of growth kinetics, plasma treatment and polymer planarising, but none were successful in lowering the WVTR. Review of this issue indicated that the achievement of good barrier layers was likely to be problematic in commercial practice due to the cost implications of adequately reducing particulate density and the need to cover deliberately non-planar surfaces and fabricated 3D structures. Conformal coverage would therefore be required to bury surface structures and to mitigate particulate issues. Studies of the remote plasma system showed that it both inherently delivered an ionised physical vapour deposition (IPVD) process and was compatible with bias re-sputtering of substrates. Accordingly, a process using RF substrate bias to conformally coat surfaces was developed to encapsulate surface particulates and seal associated permeation paths. An order of magnitude improvement in WVTR (6.7 x 10-2 g/m2/day) was measured for initial Al2O3 coatings deposited with substrate bias. The development of substrate bias to enhance conformal coverage provides significant new commercial benefit. Furthermore, conformal coverage of 5:1 aspect ratio structures have been demonstrated by alternating the substrate bias between -222 V and -267 V, with a 50 % dwell time at each voltage. Further development and optimisation of the substrate bias technique is required to fully explore the potential for further improving barrier properties and conformal coverage of high aspect ratio and other 3D structures.