One of the biggest contributors of the gas emission in smelting process of aluminium is fluoride. Major contributor to the fluoride emissions is the hydrogen fluoride (HF) gas. HF gas is formed when fluoro compounds from the cryolite react with water in the alumina. This study concerns on the rate of HF formation which is interesting due to its direct contribution to the material loss, changing of chemical composition of the bath and also the loss of current efficiency. A tunable diode laser was used to measure the HF and H2Oconcentration in the off gas during additions of alumina containing different types and amounts of water to a cryolite melt at 1000ºC. The residence time of the alumina water was varied by varying inert nitrogen gas flows through the experimental cell. These types of experiments are referred to as “melt experiment”. The water content of the alumina samples was studied by loss on ignition (LOI). In addition to the melt experiments, the HF(g) behavior inside the experimental cell was studied by introducing argon gas with 1% HF to the cell at 1000ºC when no melt was present. Finally, so called alumina water evolution experiments were conducted where the rate of alumina water released upon rapid heating was studied. Alumina was added to an empty quartz container holding 1000ºC during which the off gas analyzed with respect to H2O concentration. From the LOI result concluded that alumina B content more moisture, alumina B contains more under calcined matter compared alumina A and alumina B has higher HF formation potential than alumina A. From the alumina water evolution experiments and melt experiments found that formation of HF is very rapid. HF formation is not dependent on the nitrogen flow rate and no HF adsorption occurs as a function of flow. The amount HF generation is not necessarily flow dependent but the profile (the curve shape) and the mixing flow is dependent. Since this experimental set-up furnace can be considered as a reactor, CSTR modeling was chosen to compare the experimental data and the model.

Lack of commercial electrocatalysts and membrane electrode assemblies (MEA) which are efficient, durable and reasonably priced for proton exchange membrane water electrolysis was the inspiration of the current project.A reliable, reproducible and optimized membrane electrode assembly preparation protocol for water electrolysis was developed, with emphasis on the oxygen evolution electrode. The MEAs comprised of a Nafion® 115 membrane with commercially available 20 wt% Pt on carbon and in-house synthesized 20 wt% Ir on antimony tin-oxide, manually sprayed onto the membrane using an airbrush, and functioning as the hydrogen and oxygen evolution catalysts, respectively. A current density of 2.1 A/cm2 was obtained at a cell voltage of 1.85 V and 80 °C.In-situ electrochemical characterization such as steady state polarization and cyclic voltametry was performed on the MEAs to be able to predict performance in stationary applications. Effect of loading on cell performance at different cell voltages was studied and 0.8 mgIr/cm2 loaded MEA showed the highest current at 1.85V. Furthermore, cross section and morphology of the catalyst was studied using SEM and TEM. The catalyst layer thickness found to be from 2 to 5 µm for 0.4 to 1.0 mgIr/cm2 loadings.A theoretical one-dimensional model was proposed for current and reaction rate distribution through the catalyst layer. Since the conductivity of the catalyst measured to be higher than Nafion, model shoed at high loadings and potentials (or catalyst layer thicknesses) reaction tends to happen mostly near the membrane while at low loadings and potentials reaction rate is uniform though the layer. In addition, model was compared to experimental data and shown it is reliable for low potentials but it needs some correction for high potential due decrease in specific active area per volume by reducing thickness of the layer and not taking other factors than Tafel polarization into consideration.Introducing accelerated degradation protocol, durability of the catalyst was studied and corrected by subtracting ohmic losses due oxidation and etc. The loss for accelerated degradation found to be 0.3 mV/h. Then structure of the MEA cross-section was investigated using TEM after degradation test so major cause of the loss in performance found to be migration of Iridium nano particles into the membrane

There has been considerable interest in electrodeposition of silicon from fluoride melts on a suitable substrate for its application in thin film solar cells. The goal of this work is to produce a high purity silicon films from LiF-KF-K2SiF6 (mol %) that could be suitable for solar grade applications, and to study electrodeposition of silicon in the same melt by performing electrochemical measurements. Cyclic voltammetry was carried out both in pure melt LiF-KF and LiF-KF-K2SiF6 (0.13mol/kg) at 750oC on Ag electrode to study the reduction mechanism of fluorosilicate. The reduction mechanism was found to be mass transport diffusion controlled. The diffusion coefficient was estimated to be 1.1x10-5cm2/s from Randles-Sevcik equation.Chronoamperometry was also carried out in LiF-KF-K2SiF6 (0.13mol/kg) at 750oC on Ag electrode at different cathodic potentials and the current-time response, reduction mechanism of silicate ion was studied. It was also found again that the reduction mechanism of fluorosilicate is diffusion controlled. The diffusion coefficient was calculated to be 4.6x10-4cm2/s using the Cottrell equation. The influence of electrolytic parameters such as temperature, concentration of the electroactive species K2SiF6, and current density on the morphology and purity of the deposits and the current efficiency of the electrolytic process was studied. Effect of temperature and concentration was studied on Ag substrate and current density on both Ag and Si substrate. The influence of the substrate (silver and silicon) was also studied. Before the start of each electrodeposition experiment pre-electrolysis was carried out to remove moisture and reduce impurities concentration to ppm level. The deposits obtained were cleaned in ultrasonic bath to get rid of salt inclusions. It was observed that not all the salt inclusions were completely removed and therefore the current efficiency calculated was described to be apparent (not accurate). A few selected deposits were characterized using scanning electron microscope and energy dispersive spectroscopy, the results are shown below. It was observed that at a temperature of 800oC the deposit was dense, coherent, good adhesion to the silver substrate and with less impurities and salt inclusions but at 550oC the deposit was powdered or dendritic with high content of impurities and salt inclusions. At 5mol% of K2SiF6 the deposit consists of homogeneous structure with less impurities and salt but at 20mol% on the microstructure seems to be elongated and contain high content of impurities and salts. On silicon substrate, at current density 101.5mAcm-2 the deposit was dendritic with no grains, weakly adhered to the silicon substrate and also contains more impurities and salts but at 35.5mAcm-2 the deposit contains large grains with columnar microstructure and less impurities and salts. On silver substrate, at 83.8mAcm-2 the deposit consists of fine microstructure with high content of impurities and salts but at 42.9mAcm-2 the deposit consists of bigger and elongated with high porosity and small content of impurities and salts. For the influence of the substrate, it was observed that on sliver the deposit was insoluble with fine microstructure but on silicon it was soluble and weakly adhered with non uniform microstructure which is powdered and dendritic. The deposit obtained on both silver and silicon contains high content of impurities and salts because of the high current density applied (83.8mAcm-2 on Ag and 80.5mAcm-2 on Si).

AbstractThe aim of the present project was to investigate the possibilities of using a Zr55Cu30Ni5Al10 Bulk Metallic Glass (BMG) alloy as articulating surface in an artificial hip joint.In order for a material to be used in human body as an implant, the foremost requirement is the acceptability by the human body. The implantations should not cause diseases or other complications for the patients. Moreover, the biomaterials should possess sufficient mechanical strength, high corrosion and wear resistance in harsh body environment with varying loading conditions.There have been extensive research on the properties of stainless steel, Co-Cr-Mo alloys and Ti alloys regarding their bio-compatibility and they are currently being used as orthopedic implants, however less information is available for bulk metallic glasses. So, understanding the corrosion properties of BMGs is one of the key issues to evaluate their potential as biomaterials.In the first phase of the project there was an attempt to develop a Zr-based BMG from pure elements in a vertical resistance furnace and quenching in liquid nitrogen. Afterwards, samples were examined by X-Ray diffraction and microscopically to investigate the presence of crystalline phases. The second phase was electrochemical measurements to study the passivation behavior and the susceptibility to pitting corrosion for the crystalline Zr55Cu30Ni5Al10, amorphous Zr55Cu30Ni5Al10 BMG (received from Japan) and comparing the result with stainless steel and Co-Cr-Mo (F75). Investigations on corrosion properties were made in phosphate-buffered saline (PBS) with and without the addition of albumin fraction V, at a room temperature of 20 °C and body temperature (37°C) and in different pH values of 7.4 and 5.2. Running the experiment in lower pH shows the behavior of the implant against any probable inflation in the patient body.The last phase was to investigate the interaction between the protein and surface of materials. For this purpose, FTIR spectroscopy and Electrochemical Impedance Spectroscopy (EIS) were carried out.Keys words: Biomaterials, Bulk Metallic Glass, Crystalline Zr55Cu30Ni5Al10, Co-Cr-Mo alloy, Stainless steel, Polarization curve, FTIR, Impedance.

The oxide level of an A356 aluminium alloy will be controlled by (i) melting, casting and remelting A356 aluminium alloy for increasing the bi-films levels in the melt, and by (ii) additions of an extruded aluminium oxide rod directly to the melt. A clean A356 melt (no additions of any particulates) will be used as reference material. The melting experiments will be conducted in an electric resistance furnace under carefully controlled conditions. The three crucible experimental set-up, designed by SINTEF Materials and Chemistry, will be adopted to secure identical conditions for all three melts to be investigated. From each of the melts a series of reproducible castings will be made for tensile test measurements, as well as bend tests. The results from these tests will be statistically analysed. For the assessment of melt quality, in terms of inclusion count and bi-films index, the Porous Disc Filtration Apparatus (PoDFA) and the Reduced Pressure Test (RPT) will be adopted. For the analysis of fracture behaviour Scanning Electron Microscopy (SEM) and Energy Dispersive Spectroscopy (EDX) will be use. The dissolved hydrogen level will be measured by adopting an ALSPEC H probe.If time is given, the data generated in the present work will be compared with results previously obtained by SINTEF Materials and Chemistry for experiments conducted under similar conditions. It is believed that the results from the present study will provide a sound basis for understanding the fundamentals of the key issues involve in the production of high quality aluminium castings.