M.Sc. (Physics) / Please refer to full text to view abstract

Ph.D. / The two-step growth process, involving the selenization and sulfurization of sputter deposited CuInGa alloys has been identified as a commercially viable method to produce large area Cu(In1-xGax)(Se1-ySy)2 absorber films for solar cell application. The success of this method is however limited by insufficient control over the lattice parameters and band gap of the compound due to phase segregation associated with non-uniform Ga and S incorporation. This study provides an approach to overcome this limitation by investigating the influence of process parameters on the structural features of the Cu(In1-xGax)(Se1-ySy)2 films. In this approach, films were partially selenized in optimum H2Se/Ar flow to produce composite alloys comprising of a mixture of binary selenides (InSe, CuSe and GaSe) and at least one group I-III-VI ternary alloy. The subsequent reaction step in H2S/Ar produced homogeneous Cu(In1-xGax)(Se1-ySy)2 films. The lattice constants of the resulting films varied linearly with an increase in the S/(S+Se) ratio in accordance with Vegard’s law. The Raman spectra of the films were characterized by the presence of the A1-Se mode near 180 cm-1 and a low intensity, A1-S mode around 290 cm-1. With an increase in the S/(S+Se) ratio of the films, the FWHM of the A1-Se mode increased and its frequency shifted linearly towards that of A1-S mode. A corresponding increase in the value of the Urbach energy, attributed to an increase in chalcopyrite crystal alloy disorder, was observed from the analysis of the transmission and reflectance data. 0.45 cm2 area devices with conversion efficiencies between 12% and 15%, were fabricated from absorber layers with the (112) x-ray diffraction peak position between 27.1°and 27.2°, corresponding to the S/(S+Se) ratio of about 0.18 to 0.20. The process scale up was demonstrated by the fabrication of large area, (30 x 40) cm2 modules, with conversion efficiencies of 10%.

M.Sc. / CuInSe2 (CIS) is considered to be one of the most promising candidates for high efficiency thin film solar cells. The reaction of metallic alloys to a reactive selenium atmosphere (H2Se/Ar or elemental Se vapour) is a promising growth technique to produce CIS thin films of high crystalline quality. However, up to now, the control of the final film quality has been critically influenced by the loss of material and subsequent formation of detrimental binary phases during the high temperature selenization stages. In this study, it is shown that this phenomenon is strongly related to the selenization temperature and, in particular, the ramping procedure followed to the final selenization temperature. Metallic alloys which were selenized in H2Se/Ar at 400°C or slowly heated in 20 minutes to temperatures around 400°C were characterized by scanning electron microscopy (SEM), transmission electron microscopy (TEM) and X-ray diffraction (XRD) to have nonuniform surface morphologies, highly defected 0.8-2 !lm sized grains and to contain Cuselenide binary phases. Energy dispersive X-ray spectroscopy (EDS) analysis confirmed the generally reported sharp increase in the Cu/In atomic ratio for these classes of samples. In contrast, rapid heating (in 2 minutes) of identical metallic alloys to temperatures above 400°C, resulted in uniform, dense films with low defect density 1 !lm sized grains void of any evidence of secondary phases. X-ray fluorescence (XRF) Kal,2 measurements of metallic alloys at different stages of selenization revealed no evidence of material losses. XRF depth profiles, however, explained this discrepancy by revealing a pronounced segregation of In towards the Mo back contact when the samples were selenized at 400°C, or slowly heated to temperatures around 400°C. This segregation was dramatically reduced in films rapidly heated and selenized at temperatures above 400°C. For the purpose of comparison, metallic alloys were also reacted to elemental Se vapour. The structural features (grain size and preferred orientation) ofthese films differed significantly from those selenized under similar conditions in H2Se/Ar. The results from this study, including photoluminescence (PL) measurements obtained from these films, were used to affect the fabrication of CIS absorbers with excellent material properties and solar cell devices with moderate conversion efficiencies.

No description available.

No description available.

No description available.

In order to investigate the influence of parasitic resistances, saturation current and diode ideality factor on the performance of photovoltaic devices, parameter extraction routines employing the standard iteration (SI) method and the particle swarm optimization (PSO) method were developed to extract the series resistance, shunt resistance, saturation current and ideality factor from the I-V characteristics of solar cells and PV modules. The well-known one- and two-diode models were used to describe the behavior of the I-V curve and the parameters of the models were determined by approximation and iteration techniques. The SI and the PSO extraction programmes were used to assess the suitability of the one- and the two-diode solar cell models in describing the I-V characteristics of mono- and multicrystalline silicon solar cells, CISS- and CIGSS-based solar cells. This exercise revealed that the two-diode model provides more information regarding the different processes involved in solar cell operation. Between the two methods developed, the PSO method is faster, yielded fitted curves with lower standard deviation of residuals and, therefore, was the preferred extraction method. The PSO method was then used to extract the device parameters of CISS-based solar cells with the CISS layer selenized under different selenization process conditions and CIGSS-based solar cells with varying i-ZnO layer thickness. For the CISS-based solar cells, the detrimental effect of parasitic resistances on device performance increased when the temperature and duration of the selenization process was increased. For the CIGSS-based devices, photogeneration improved with increasing i-ZnO layer thickness. At high forward bias, bulk recombination and/or tunneling-assisted recombination were the dominant processes affecting the I-V characteristics of the devices. v Lastly, device and performance parameters of mono-, multicrystalline silicon and CIS modules derived from I-V characteristics obtained under dark and illuminated conditions were analyzed considering the effects of temperature on the performance of the devices. Results showed that the effects of parasitic resistances are greater under illumination and, under outdoor conditions, the values further declined due to increasing temperature. The saturation current and ideality factor also increased under outdoor conditions which suggest increased recombination and, coupled with the adverse effects of parasitic resistances, these factors result in lower FF and lower maximum power point. Analysis performed on crystalline silicon and thin film devices utilized in this study revealed that parameter extraction from I-V characteristics of photovoltaic devices and, in particular, the implementation of PSO in solar cell device parameter extraction developed in this work is a useful characterization technique.

In this study a presentation of the technology behind dye-sensitized solar cells, their design as well as the role of the different parts of the cell. The characterization of the cell is divided into four sections namely: the characterization of the paste required to make the TiO2 film and its optical properties using SEM-EDX and XRD analytical techniques; Analysis of the various absorptions of three Ru-based dyes using UV-Vis spectroscopy, Photoluminescence and Fourier Transform Infra-Red spectroscopy; the characterization and the analyses of the entire cell using Electrochemical Impedance Spectroscopy. The nine cells were prepared by examining RuL2(CN)2 , RuL2(NCS)2 or N3 dye and RuL2(NCS)2 TBA+ or N719 dye. [L = 2,2'-bipyridyl-4,4'-dicarboxylic acid ;TBA = tetra-butyl ammonium] were combined with three electrolytes namely: Z–150 , AN–50 and PN–50. The Iodolyte PN–50 is an iodide based low viscosity electrolyte with 50 mM of tri-iodide dissolved in a solvent called propionitrile (PN). The Iodolyte AN–50 is an iodide based low viscosity electrolyte with 50 mM of tri-iodide dissolved in a solvent called acetonitrile (AN). The Iodolyte Z–150 is an iodide based low viscosity electrolyte with 150 mM of tri-iodide dissolved in a solvent called 3-methoxypropionitrile (MPN) and with additives such an ionic liquid, malkylbenziimidazole and guanidine thiocyanate. A solar simulator was utilized with which the standard solar irradiation can be created in laboratory conditions. The fill factors as well as overall performance efficiencies of the these cells are quite low < 1.0%,.

Thin-film photovoltaics are of great interest due to decreased manufacturing costs, improved environmental sustainability and the potential for flexible, semi-transparent, and light-weight modules. The scientific literature contains a plethora of work incorporating wavelength scale nanostructures within thin-film solar cells to increase power conversion efficiency by trapping light inside solar cell absorbing layers. One category of nanostructures, namely plasmonic nanoparticles, theoretically show great promise for their light-trapping abilities but experimental success has been limited. In this work, solar cells were designed and fabricated to incorporate multiple light-trapping mechanisms, including optical cavity resonances, waveguide mode excitation, and plasmonic effects. Due to our novel design considerations, we demonstrate a 33% increase in Jsc originating from plasmon-based enhancement mechanisms. The experimental results are complemented and confirmed by well-matching simulations which are used to further investigate the light-trapping mechanisms. The concepts demonstrated in this work can be directly translated to next-generation transition metal dichalcogenide photovoltaic devices. / Graduate / 2018-12-14

Over the past decade, intensive academic and commercial interests have been paid on compounds possessing photochemical properties, namely for their preparation, chemical properties, high efficiency and potential low-cost. Compounds having intense photochemical properties gained great interest due to wide range of potential applications. The sensitizers are one of the key components for high power-conversion efficiency in the dye sensitized solar cells (DSSCs). They are the core components in the organic light-emitting devices (OLEDs) due to their ability to emit light with the wavelengths largely red- shifted from their absorption wavelength. Ruthenium based sensitizers have been tagged “molecular light switches” because, although the fluorescence of these complexes in aqueous solutions is negligible, it increases of greater than 10000 fold in the presence of DNA. Many polypyridyl and dipyrido phenazine ruthenium complexes have achieved high power conversion efficiencies and therefore are of practical interest. Several research groups stated that the dipyrido phenazine ligand may be thought of as comprising two components: a bipyridyl unit and a phenazine unit. These two subunits behave essentially separately, with many molecular orbitals being localised over only one subunit and a redox properties of central phenazine moiety in the dipyrido phenazine ligand are important for the photochemical applications. Therefore a phenazine ligand was selected as a model for the present investigation. The chemistry of phenazine ligand is mostly limited to the late transition metal and f - element complexes. Our laboratory has a rich backgroung in the aluminum and early transition metal chemistry. The aluminum chemistry and early transition metal chemistry are of great interest since aluminum and early transition metal complexes are environmentally friendlier and cheaper than the late transition metal compounds. Another drawback of the ruthenium-based sensitizers is the lack of absorption in the red region of the visible spectrum, and also low molar extinction coefficients. An essential requirement for efficient conversion of solar energy is the good spectral match of the sensitizer absorption to the emission spectrum of solar radiation. In this regard, the ruthenium sensitizers’ spectral response in the lower energy regions is not sufficient. The current project has three parts. In the first part we collected and reviewed known literature regarding the certain classes of non-innocent ligands containing the six-membered carbon- nitrogen heterocycles and regarding the ligands potentially important for the photochemical applications. We also reviewed all available to the data information about the complexes supported by the phenazine ligand. In the second part we have investigated interaction of alkylaluminum compounds and phenazine and observed reduction of phenazine accompanied by formation of dialuminum cage type compounds containing two formally mononegative phenazine ligand. The derivatization of phenazine has been also observed. It resulted in formation of compounds having a stable organic radical. In a third part of our project we have explored interaction of phenazine or thiophenazine with the alkylaluminum compounds and chromium dichloride. The reaction in the three component system resulted in reduction of phenazine ligand and lead to the heterometallic Cr(II) - aluminum complexes containing a formally dinegative phenazine or thiophenazine ligands. When a large excess of triethylaluminum was taken, reduction of phenazine and chromium has been observed leading to the heterometallic multinuclear Cr(I) - aluminum complex containing a formally dinegative phenazine ligands and two chromium atoms in one complex in the rare oxidation state one.