The solubility of divalent iron oxide in cryolite-based melts was studied. Both electrochemical and chemical techniques were employed. To ensure that only divalent iron was present in solution, the melt was contained in an iron crucible under an atmosphere of argon. The experimental work included investigation of the solubility as a function of alumina concentration, temperature and cryolite ratio (CR = NaF/AlF3 molar ratio). The solubility at 1020 ºC was found to decrease from 4.17 wt% Fe in cryolite to 0.32 wt% Fe in cryolite saturated with alumina. FeO and FeAl2O4 were found to coexist as solid phases in equilibrium with the melt at 5.03 wt% Al2O3; the former being the stable solid phase below this concentration and the latter at high alumina concentrations. The standard Gibbs energy of formation for FeAl2O4 from its oxide components at 1020 ºC was determined to be -(17.6 ± 0.5) kJ mol-1. The solubility of FeAl2O4 was found to increase from 0.25 wt% Fe at 981 ºC to 0.36 wt% at 1050 ºC in alumina-saturated melts. By assuming Henrian behaviour, the apparent partial molar enthalpy of dissolution of FeAl2O4 was found to be (64.8 2.5) kJ mol-1. Experiments involving varying cryolite ratio in alumina-saturated melts at 1020 ºC showed a maximum solubility of 0.62 wt% Fe at a cryolite ratio of five. Modelling indicated that divalent iron species were present as NaFeF3 in acidic melts (CR < 3), while Na3FeF5 and/or Na4FeF6 dominated in a basic environment (CR > 3). The solubility of TiO2 in cryolite-alumina melts at 1020 ºC was measured. The analytical data showed that the titanium solubility decreased with increasing total oxide concentration, up to a concentration of ~3.5 wt% O, while it increased at higher concentrations. The solubility was found to be 3.1 wt% Ti and 2.7 wt% Ti, respectively, in cryolite and in alumina-saturated melts. Modelling indicated that the most probable titanium species are TiO2+ and TiO32-, which coexist in the solution; the former dominating at low alumina concentrations and the latter at high alumina concentrations. Unknown amounts of fluoride may also be associated with the titanium atoms. Determination of the solubility of TiO2 in alumina-saturated melts as a function of temperature showed that the solubility increased from 1.9 wt% Ti at 975 ºC to 2.8 wt% Ti at 1035 ºC. The apparent partial molar enthalpy of dissolution of TiO2 was found to be (88.3 ± 4.1) kJ mol-1, provided that Henry’s law holds. The electrochemistry of divalent iron in cryolite-based melts was investigated by voltammetry, chronopotentiometry and chronoamperometry. A working electrode of copper was found to be best suited for the study of the reduction of Fe(II), while gold and platinum gave the best results under oxidising conditions. The reduction of Fe(II) ions was found to be diffusion controlled. The number of electrons involved was determined to be two. A discrepancy was observed between the diffusion coefficients obtained by the different techniques. The diffusion coefficient of Fe(II) in alumina-saturated melts at 1020 ºC was found to be DFe(II) = 3.0 x10-5 cm2 s-1 by voltammetry. Experiments performed in an electrolyte with industrial composition at ~970 ºC gave a slightly higher value for the diffusion coefficient. The oxidation of Fe(II) on a gold or a platinum wire electrode showed that the process was diffusion controlled, involving one electron. The reversible potential for the redox couple Fe(III)/Fe(II) was found to be more cathodic than the reversible potential for the oxygen evolution by 350 to 400 mV, depending on the solvent composition and on the temperature. The electrochemistry of TiO2 in cryolite-alumina melts was studied by voltammetry. The deposition of titanium on tungsten was found to be a three-electron diffusion controlled process. The deposition peak increased with increasing titanium concentration. In alumina-saturated melts two waves were observed prior to the titanium deposition. The potential difference between the cathodic wave closest to the deposition peak and its corresponding oxidation peak indicated a diffusion controlled process that involved a one-electron charge transfer. However, in cryolite melts a single wave was observed prior to the titanium deposition. It is suggested that these cathodic waves might have been caused by underpotential deposition of titanium, and subsequent alloying with tungsten. It cannot be ruled out that redox reactions take place between tetravalent titanium and the titanium alloyed with tungsten, thereby forming trivalent titanium prior to the metal deposition. In order to determine thermodynamic properties of FeAl2O4, a solid electrolyte galvanic cell was used. Cryolite was present in the half-cell containing FeAl2O4 to ensure that alumina of the alpha modification was in equilibrium with FeAl2O4. An oxygen ion conducting yttria-stabilised zirconia tube served as the solid electrolyte. The EMF was measured in the rage 1245 to 1343 K. By using literature data at higher temperatures, thermodynamic properties for the reaction Fe(s) + ½O2(g) + Al2O3(s,α) = FeAl2O4(s) could be calculated, i.e. ΔHº1600K = –(270615 ± 1387) J mol-1 and ΔSº1600K = -(56.759 ± 0.856) J K-1 mol-1. New thermodynamic data for FeAl2O4 were also calculated, and a predominance area diagram for solid iron phases at 1293 K was constructed. The standard potential of the redox couple Fe(III)/Fe(II) as a function of the alumina content was derived from the solubility data of Fe(II) obtained in the present work and literature data for Fe(III). When the standard potentials are put into context of the Hall-Héroult process, the results indicate that neither the CO2/CO anode gas nor the carbon anode itself can oxidise Fe(II) to Fe(III). The mass transfer of the impurities Fe, Si and Ti between bath and aluminium in industrial Hall-Héroult cells was investigated. The experiments were performed in several types of cells with prebaked anodes. The impurities were added to the bath in the form of Fe2O3, SiO2 and TiO2. Bath and metal samples were collected periodically before and after the addition was made. With the criterion that the mass transport was diffusion controlled, a model involving first order reaction kinetics was used to calculate the mass transfer coefficients for transfer into the metal phase. Large scatter were observed in the obtained mass transfer coefficients, but the general trend seemed to be kFe > kSi > kTi. By averaging the data obtained, it was found: kFe = (10 ± 3) x 10-6 m s-1, kSi = (7 ± 3) x 10-6 m s-1, and kTi = (5 ± 2) x 10-6 m s-1.

Welding is the main joining method used in shipbuilding. A multi-electrode submerged arc welding is usually applied to long joints of massive components since the early 1950s. The problem of welding stresses and deformations arose almost simultaneously with the introduction of welding as a joining method. In 1892 Slavianov, in his works, mentioned about "harmful stresses in metal" appearing during welding. Despite tremendous development in welding technology since that time, this problem is still present. This dissertation presents the development of an experimental, numerical and analytical approach to the analysis of weldability. Focus is placed on the investigation of transient and residual welding stresses and distortions in thick-walled ship hull structures. The investigation mostly relies on the finite element analysis. Some recommendations about mitigation techniques for reducing their negative influence on welded structure efficiency are given. A number of simulation systems for the sequentially coupled thermo-mechanical analysis of welding process, taking into account the welding conditions during the butt welding of thick-walled steel panels, are developed. Two welding techniques are investigated. The three-electrode one-pass welding process and the one-electrode multipass process are compared in terms of the joint characteristics and transient stresses and deformations. The accuracy of each finite element models is evaluated based on experimental results and the results of the analytical solution. Several techniques allowing to reduce residual stresses and deformations are investigated. Both, obtained results and governing phenomena, are studied and explained based on physical principles. The first part of the thesis is a significant part of the work. Its purpose is a deep analysis of the thermo-mechanics of welding allowing better understanding of the complicated phenomena.

This dissertation deals with the unintended sideeffects of utilizing resources and causing environmental load. The intention of this report is to contribute to a leap improvement in enviornmental performance of industrial products. The results of the dissertation are directed towards companies that want to eliminate, or at best reduce the unintended side effects of their activities and industrial products. The results are directed to the research community as a contribution to research on the effect of a shift from product-based to function-based economic system, and the potential improvement in environmental performance this may imply. The main contribution of this dissertation is a strategy for developing product service combinations and thereby closing material loops of industrial products. The result is based upon product development models, creative tools and performance measurement combined with the resolution of the basic physical contradiction into a four step strategy for a long term shift of a business model based on product service combinations instead of products only. The fourstep strategy involves traditional product development models with focus on holistic lifespan thinking. Furthermore, emphasize is made to the need to include the creative tools to conquer psychological inertia and barriers for leap improvements in environmental performance. The strategy is mainly based upon logical resoning and thought experiments and combined with observations in industry. It is difficult to give a final proof of the usefulness of the strategy, due to the time needed to perform such a strategic shift in the core activities of a business.This research has furthermore contributed to • a clarification of the source of environmental load during the meetings between product and product life system where any action to reduce the original load will result in load of its own which means that the fundamental cause of load is still not resolved. Any attempt to remove the fundamental cause of side-effect is a true physical contradiction which violates the second law of thermodynamics. • a clearer understanding of the need not only to integrate the different disciplines across a company’s operations, but also integrate along the value chain to capture the valuable information which arise during meetings between product and product life systems. • a deeper understanding of the need to include creative techniques into the early phases of product development proejcts where dispositions for subsequent project phases and subsequent product life phases are disposed and locked against subsequent major modifications. Creative techniques contribute to the break out of existing mind patterns and contribute to the creation of solutions which reduce the business-as-usual pattern which acts as a barrier to leap improvement in environmental performance • a clarification of the need to identify and develop the company’s environmental perspective. The need to develop strategies for improvingenvironmental performance of company internal processes, product systems and systemic networks of actors along the value chain is vital to achieve leap improvements in environemntal performance. • a deeper understanding of the barriers to overcome to transform products-based business to service-based business.

This doctoral thesis is concerned with pitting corrosion in super martensitic stainless steel (SMSS) weldments in slightly sour service. Thee main objective with the present thesis has been to find out why pitting corrosion occurs in the heat affected zone (HAZ) at ambient rather than at elevated temperatures and how the corrosion mechanism depends on the welding process. The thesis is divided into six parts. Part I gives a general introduction to martensitic stainless steels, focusing on the metallurgical and corrosion properties. Part II is concerned with the conditions for initiating of pitting corrosion in two SMSS weldments. Optical microscopy in combination with sputtered neutral mass spectrometry has been used to examine the conditions, and it is shown that the corrosion resistance depends mainly on the nature of the surface oxide and less on the underlying microstructure. In the absence of H2S the initiation is associated with spalling of the iron-enriched chromium oxide within a narrow region approximately 6mm from the fusion boundary, where the contaminated oxide layer is thinnest. As H2S is introduced, the region close to the fusion boundary become susceptible to localised corrosion because of the more extensive metal oxidation. Thus, the high temperature oxidation and iron oxide scale formation accompanying the welding operation appears to be the main factor affecting the pitting corrosion initiation in SMSS weldments. In Part III, a comparative study of the dissolution and repassivation behaviour of a Fe-12.3Cr-6.5Ni-2.6Mo SMSS alloy and other stainless grades has been carried out. The SMSS alloy is characterised by means of the so-called “artificial pit technique”, and a diffusion model has been developed and employed for calculation of the pit surface concentration of dissolved species during the potential step experiments. For concentrations close to the saturation level, the dissolution kinetics are adequately described by a Tafel slope of approximately 57mV/dec and a current density of 0.5mA/cm2 at –300mV SCE. On the other hand, repassivation of the active pit surface occurs when the concentration of the dissolved species drops below 30% of the saturation value. Based on a comparison with relevant literature data it is concluded that the observed response of the SMSS to localised corrosion is similar to that reported for other high-alloyed steels. This result is to be expected if the dissolution and repassivation kinetics are controlled by the content of Cr, Ni and Mo in the parent material. In Part IV, the specific effects of H2S on the pitting corrosion behaviour of a Fe-12.3Cr-6.5Ni-2.6Mo SMSS alloy have been examined by means of the “artificial pit technique”. Addition of a critical amount of H2S is found to hinder repassivation and accelerate active dissolution in the SMSS pit within the stability domain of the adsorbed sulphur. Above the redox potential of H2S/S, the effect of H2S is found to cease, whereby dissolution and repassivation kinetics similar to H2S-free solutions are observed. Activated pits show no signs of repassivation, not even at low surface concentration of dissolved species. It is concluded that the resistance to pitting corrosion depends both on the potential and the H2S concentration and the effect of H2S is most significant at low potentials. Part V is concerned with the occurrence and appearance of pitting corrosion in SMSS girth welds in the presence of H2S. Galvanostatic and potentiodynamic measurements were carried out at ambient temperature and pressure, using both a deaerated solution and a 1mM H2S solution. In conjunction with this study, a descriptive pitting corrosion model is proposed to explain the effect of the root surface condition and the role of H2S. The oxidized pipe surface close to the fusion boundary is found to be extremely vulnerable to pitting corrosion in the presence of H2S, while the same zone appears to behave inert in the deaerated solution. In the latter case the oxidised surfaces were found to repassivate. In the H2S solution, localized corrosion occurred in the most oxidized region of the HAZ, about 0 to 3mm from the fusion boundary. Moreover, it was found that a modification of the as-welded root surface by grinding hindered any initiation of pitting corrosion in this region, again highlighting the important role which high temperature oxidation and oxide scale formation play in controlling the corrosion properties of SMSS weldments. Finally, in Part VI of the thesis, the practical implications of the above findings for choice and development of relevant corrosion test procedures have been considered. A key issue here is the effect of pre-exposure on the corrosion resistance of SMSS weldments in the presence of H2S. Pre-exposures have been simulated in test solutions resembling the conditions in a flowline. It is shown that the quality (i.e. passivity) of SMSS root surfaces can be largely improved by the use of an appropriate pre-exposure procedure. It is therefore recommended that all samples in the future are pre-exposed to an H2S-free solution before they are subjected to actual corrosion testing in solutions containing H2S to avoid pitting corrosion in the oxidized part of HAZ. This is deemed to be important in order to achieve realistic results.

The processing of aluminium alloys from casting to end product is associated with a large number of metallurgical phenomena. In order to further improve and optimise process routes and alloys, a thorough understanding of the thermomechanical treatments by experimental observations and physically based modelling is necessary. In part A of this thesis the work hardening behaviour of non-heat treatable alloys is followed up to large strains. The evolution in strength, microstructure and texture during cold rolling is analysed in commercially pure aluminium, AlMg alloys and AlMn alloys, covering both solid solution strengthening, particle strengthening, grain size effects and effects of impurities. A recently developed model for work hardening is applied to these alloys and discussed in relation to the experimental observations. The model successfully predicts the effects of Mg in solid solution, and is used to explain the effects of nonshearable particles and grain size on the strength and work hardening in stage II-III. Mn, Fe and Si in solid solution are suggested to create solute clusters that increase the strength significantly. At higher strains, stage IV, an unexpectedly low work hardening rate of high-Mg alloys is ascribed to a different storage pattern of dislocation and an increased amount of shear bands. High-resolution EBSD in FE-SEM is demonstrated to be a promising tool for substructure characterisation of cold rolled alloys. In part B the mechanical anisotropy of flat extruded profiles of heat-treatable alloys is investigated. Two recrystallised alloys, AA6063 and 7030, and two non-recrystallised alloys, AA6082 and 7108, are tested in uniaxial tension in the solution heat treated condition, exhibiting strong directionality of yield stress, work hardening rate, elongation and r-value. The initial textures are very strong, and the anisotropy is analysed in terms of the Taylor model. The texture is found to be responsible for most of the strength anisotropy in the non-recrystallised alloys, while other sources of anisotropy must be present in the recrystallised alloys. Variations in the deformation structure development, indicating different slip activity in different directions, are believed to be partly responsible. The Taylor RC model predicts the r-values very well in all alloys, and a number of active slip systems in each position between two and three is assumed to be realistic for all alloys and directions.

Pitches are used on a large scale in the manufacture of carbon anodes for the production of primary aluminium. The role of the pitch is to act as a binder between the petroleum coke grains. The structure of the carbonized pitch binder (pitch coke) has an important impact on the overall performance of the anode. Even though the binder pitch is the minor constituent in an anode, it is impossible to make a good quality anode without a good quality binder pitch. Pitch is an extremely complex mixture of numerous, essentially aromatic and heterocyclic compounds derived from pyrolysis of organic material or tar distillation. Upon heat treatment pitches form cokes in relatively high yields. Physical and chemical properties of the anode such as mechanical strength, electrical resistivity, thermal conductivity and resistance towards oxidation by air and CO2 are dependent on the structure of the aggregate material as well as the carbonized binder pitch. The properties of the pitch coke is in turn mainly dependent on the chemical characteristics of the parent pitch. Coal-tar pitch is the preferred choice of binder material in anode manufacture today. However, the availability of high quality coal-tar is in decline and at least partial replacement by alternative binder sources will become increasingly important in the future. Due to environmental regulations, petroleum pitches are interesting as they generally have lower PAH emissions than coal-tar pitches during baking. Blends of coal-tar pitches and petroleum pitches are in use today on an industrial scale. The aluminium industry must be prepared to meet the challenges involved in adapting binder pitches from new sources which may be of inferior quality to the pitches available on the market today. An increased understanding of the processes involved in the transformation of a pitch into a coke and the link between raw material composition and properties and the final artifact is thus highly relevant. Traditionally, the suitability of a binder pitch for use in anodes, has been defined from parameters like softening point, insolubility in toluene (TI) and quinoline (QI), coke yield, H/C atomic ratio, ash content and density. Although these parameters, which are mostly empirical in nature, give an indication of the pitch quality, more information on the chemical characteristics and carbonization behavior of pitches is certainly valuable. The present work aims to describe and explain the link between “classical” pitch properties, hydrogen transfer properties, information derived from NMR spectroscopy and the structure of the carbonized binder pitch. Coal-tar and petroleum pitches pass through a fluid stage during carbonization. In the early stages of carbonization, free radicals are formed due to thermal rupture of C-C and C-H bonds in reactive components. Polymerization occurs mainly via a free radical mechanism leading to molecular size enlargement (aromatic growth)and the formation of oligomeric systems (mesogens). If the intermolecular reactivity of the pitch constituents is too high, extensive cross-linking and a rapid transformation of pitch molecules through polymerization will occur at a relatively low temperature. In this case, either mesophase will not be formed or the growth and coalescence of mesophase will take place under low fluidity/high viscosity conditions leading to a premature solidification of the pyrolysis system. An isotropic coke or a pitch coke of small optical domains will then be formed. On the other hand, if the pitch has a low thermal reactivity, aromatic growth is constrained and the mesogens will have sufficient mobility to stack parallel to each other and establish a liquid crystal system (mesophase). The growth and coalescence of mesophase take place at a higher temperature where the viscosity of the pyrolysis system is at a low level. Eventually, the system will solidify and an anisotropic coke of large well-developed optical domains is formed. In particular, the presence of alkyl side chains and oxygenated functional groups are considered to lead to an increased thermal reactivity. If free radicals formed by thermal rupture of bonds in reactive pitch species can be stabilized by hydrogen transfer from within the system, extensive cross-linking at a too early stage is prevented. The initiation, growth and coalescence of mesophase are facilitated and consequently a coke of large well-developed optical domains is formed. Hydroaromatic rings and naphthenic rings in hydroaromatic species are considered to be principal hydrogen donor groups. Oxygen acceptor sites are believed to deplete the supply of donatable hydrogen and leave radicals free to recombine. The thermal reactivity of a pitch is thus dependent on both the amount of reactive species and the ability of the pitch to stabilize free radicals by hydrogen transfer. In the present work, the subject of study was five coal-tar pitches and four petroleum pitches. In addition, a QI-free coal-tar pitch supplied by GrafTech International was studied. The pitches were characterized by 1H NMR and 13C NMR spectroscopy, hydrogen transfer properties, elemental analysis and the release of volatiles during carbonization. In addition, the pitches were characterized by more “traditional” pitch parameters like insolubility in quinoline (QI), insolubility in toluene (TI), softening point and coking value. The structure of the carbonized pitches was examined by optical microscopy and X-ray diffraction. The hydrogen transfer properties of the pitches were evaluated from their ability to donate hydrogen to an acceptor compound, anthracene, or abstract hydrogen from a donor compound, 1,2,3,4-tetrahydronaphthalene (tetralin). A mixture of pitch and anthracene or tetralin was heat treated in sealed glass tubes filled with argon gas at 400 ºC. Two different heat treatment procedures were tested. In the first, the sample was kept at 400 ºC for 8 hours while in the second, the sample was heated at a rate of 5 ºC/min to 400 ºC with no soaking time. The major hydrogenated products from the reaction between anthracene and pitch were 9,10-dihydroanthracene (DHA) and 1,2,3,4-tetrahydroanthracene (THA). After the reaction, the semi-coke residue was dissolved in carbon disulphide and analyzed by gas chromatography. The hydrogen donor ability (HDa) was calculated from the amounts of DHA and THA formed and expressed as milligrams of hydrogen transferred to anthracene per gram of pitch. For the hydrogen donor ability test, the less severe heat treatment (5ºC/min to 400 ºC, no soaking time) was found to be the most appropriate. The reaction between tetralin and pitch gave one major dehydrogenated product, naphthalene. The hydrogen acceptor ability (HAa) was calculated from the ratio of naphthalene to tetralin as determined by gas chromatography and expressed as milligrams of hydrogen transferred per gram of pitch. For the acceptor ability test, the heat treatment at 400 ºC with 8 hours soaking time was found to be the most appropriate. The release of volatiles during carbonization was studied by thermogravimetric analysis. The amount of volatiles released between 300 and 500 ºC (VM300-500)relative to the total amount of volatiles released at 1000 ºC was selected as a parameter reflecting the thermal behavior of pitches during the critical stages of carbonization. Carbonization of pitches was performed under inert gas pressure (15 bar) and the green cokes obtained at 550 ºC were studied by optical microscopy. Computerized image analysis was performed to quantify the optical texture. The output parameters from the image analysis were the mosaic index, which is a measure of the optical domain size, and the fiber index, which is a measure of the parallel alignment of optical domains. The green cokes were further heat treated to 1150 ºC and the microstructure of the resulting calcined pitch cokes was characterized by X-ray diffraction. The carbon disulphide soluble part of the pitches was investigated by 1H NMR and 13C NMR spectroscopy. Results from elemental analysis of the pitches were used in conjunction with the results obtained from the NMR spectroscopy. The main objective of the NMR analysis was to identify and quantify structures in the pitch which are considered either to increase or decrease the thermal reactivity. The coal-tar pitches were as expected found to be more aromatic than the pitches of petroleum origin. A relationship was found between the aromaticity of the pitches and the H/C atomic ratio as determined from elemental analysis. Elemental analysis is a rapid and convenient method to estimate the aromaticity of pitches. Due to a more hydroaromatic structure, the petroleum pitches were in general found to have a higher estimated concentration of donatable hydrogen which will suppress intermolecular reactivity. However, the petroleum pitches also had a high concentration of alkyl side chains which are generally believed to give increased thermal reactivity. Carbon connected to oxygen could not be distinguished in the NMR spectra. Pitch constituents containing heteroatoms are generally concentrated in the heavier pitch fractions which may not be soluble in carbon disulphide. This could be an explanation for the failure in the detection of aromatic carbon connected to heteroatoms. However,the oxygen content was determined by elemental analysis. The pitches could be distinguished due to their ability to donate hydrogen to anthracene or abstract hydrogen from tetralin. The hydrogen donor ability was not found to correlate with the concentration of donatable hydrogen (NMR) which might have been expected. A likely explanation for this apparent inconsistency is that potential donatable hydrogen in reactive pitches will be preferentially consumed by free radicals and oxygenated acceptor sites instead of being transferred to anthracene. A correlation between the hydrogen donor (HDa) and acceptor ability (HAa) was not found. This indicates that the two parameters represent two separate properties where both are linked to the thermal reactivity of the pitch. The ratio between the hydrogen donor and acceptor ability, HDa/HAa, was used as a parameter reflecting the thermal reactivity of pitches. Pitches which exhibit a high HDa/HAa ratio (low thermal reactivity) are expected to form an anisotropic coke of large optical domains. On the other hand, pitches with a relatively low HDa/HAa ratio are expected to have a high thermal reactivity and form a more isotropic (small optical domains) coke. Despite the higher concentration of donatable hydrogen, the petroleum pitches were not generally considered to have a lower thermal reactivity than the coal-tar pitches expressed by the HDa/HAa ratio. The processes taking place during thermal treatment of pitches are reflected in the release of volatiles. A correlation was observed between the HDa/HAa ratio and the relative amount of volatiles released between 300 and 500 ºC (VM300-500). Thermally reactive pitches exhibiting a low HDa/HAa ratio will have a high activity at low temperatures and release low boiling point molecules and fragmentation species. If on the other hand the pitch has a low thermal reactivity, fragmentation species will be stabilized by hydrogen transfer and retained in the pyrolysis system. The resulting thermally stable molecules of relatively low molecular weight may then act as solvating vehicles maintaining a low viscosity in the system and may also be important as hydrogen shuttling agents. When the system has reached a critical stage for mesophase growth and coalescence, these smaller thermally stable molecules (non-mesogens) are eventually released at higher temperatures. The petroleum pitches developed cokes of relatively large optical domains (coarse mosaic). A correlation was observed between the HDa/HAa ratio and the mosaic index (size of optical texture) for the petroleum pitches. As expected, a high thermal reactivity (low HDa/HAa ratio) resulted in a pitch coke of small optical domains (high mosaic index). The HDa/HAa ratio was, however, not successful in predicting the size of optical texture in the cokes obtained from the coal-tar pitches. This was mainly due to the influence of QI material on the pitch coke structure. It is recognized that particulate matter (primary QI material) hinders the growth and coalescence of mesophase. This was found for the coal-tar pitches. Scanning electron (SEM) and polarized light microscopy images taken at a high magnification revealed how the QI particles were arranged and clustered around smaller anisotropic domains. The detrimental effect of QI material on the development of anisotropic texture in the resulting coke was demonstrated by comparing the structure of the coke obtained from a QI-free coal-tar pitch and a coal-tar pitch containing QI. The QI-free pitch developed a coke of large optical domains whereas the coke obtained from the pitch containing QI material had mainly a fine mosaic texture (small optical domains). However, some large anisotropic domains were present in between the QI clusters. It is also not to be excluded that the QI fraction not only acts physically by obstructing the growth and coalescence of mesophase but may also be chemically active. Findings indicate that the oxygen is concentrated in the QI fraction. Solid QI particles with oxygenated functional groups or heteroatomic structures containing oxygen, which due to their large size are insoluble in quinoline, may act as acceptor sites for hydrogen thus increasing the thermal reactivity. The average coherent stacks of the calcined (1150 ºC) pitch cokes was found to consist of between 7 and 8 graphene layers (Lc divided by d002). The average crystallite size (Lc) was fairly similar for all the calcined pitch cokes but significant differences were found. The coal-tar pitches generally developed cokes of slightly higher average crystallite sizes than the pitches of petroleum origin. The microstructure of the coal-tar pitch cokes is probably influenced by the amount and nature of the QI fraction. For the petroleum pitches there was a tendency that a high average crystallite size was connected to a more well-developed structure (larger domains) at the green coke stage. The evaluation of hydrogen donor and acceptor abilities provides a rapid and relatively simple method to differentiate pitches which can be linked to the development of structure during carbonization. These properties thus reflect the thermal reactivity of pitches and can be connected to the release of volatiles during pyrolysis. However, for coal-tar pitches the QI content was found to be the most influential factor on the development of optical texture and must be considered in addition to the hydrogen transfer properties. Considerations on thermal reactivity from NMR spectroscopy and elemental analysis were found to generally support the results from the hydrogen donor and acceptor ability tests.

The main objectives of the thesis work were: (1), to study the oxidation of methanol and formic acid on platinum electrodes by employing conventional and advanced electrochemical methods, and (2), to develop membrane electrode assemblies based on polybenzimidazole membranes that can be used in fuel cells up to 200 °C. D.c. voltammetry and a.c. voltammetry studies of methanol and formic acid on polycrystalline platinum in sulphuric acid electrolyte were performed to determine the mechanism and kinetics of the oxidation reactions. A combined potential step and fast cyclic voltammetry experiment was employed to investigate the time dependence primarily of methanol oxidation on platinum. Charge measurements clearly demonstrated the existence of a parallel path at low potentials and short times without formation of adsorbed CO. Furthermore, experimental results showed that only the serial path, via adsorbed CO, exists during continuous cycling, with the first step being diffusion controlled dissociative adsorption of methanol directly from the bulk electrolyte. The saturation charge of adsorbed CO derived from methanol was found to be significantly lower than CO derived from formic acid or dissolved CO. This was attributed to the site requirements of the dehydrogenation steps, and possibly different compositions of linear, bridged or multiply bonded CO. The coverage of adsorbed CO from formic acid decreased significantly at potentials just outside of the hydrogen region (0.35 V vs. RHE), while it did not start to decrease significantly until about 0.6 V vs. RHE for methanol. Adsorbed CO from dissolved CO rapidly oxidized at potentials above about 0.75 V due to formation of platinum oxide. Data from a.c. voltammograms from 0.5 Hz up to 30 kHz were assembled into electrochemical impedance spectra (EIS) and analyzed using equivalent circuits. The main advantages of collecting EIS spectra from a.c. voltammetry experiments are the ability to directly correlate the impedance spectra with features in the corresponding d.c. voltammograms, and the ability to investigate conditions with partially covered surfaces that are inaccessible in steady-state measurements. A variety of spectral types were observed, and for methanol these showed only a single adsorption relaxation aside from the double-layer/charge-transfer relaxation, though some structure in the phase of the latter relaxation hints at another process. The charge-transfer resistance showed Tafel behaviour for potentials in the rising part of the oxidation peak consistent with a one-electron process in the rate-determining step. The rate limiting step was proposed to be the electrochemical reaction between adsorbed CO and OH at the edge of islands of OH, with competition between OH and CO adsorption for the released reaction sites. Only a single adsorption relaxation in methanol oxidation was observed, implying that only one single coverage is required to describe the state of the surface and the kinetics. It was assumed that this single coverage is that of OH, and all the surface not covered with OH is covered with CO so that the coverage of CO is not an independent variable. Inductive behaviour and negative relaxation times in the methanol oxidation were attributed to nucleation and growth behaviour. Linear voltammetry reversal and sweep-hold experiments also indicated nucleation-growth-collision behaviour in distinct potential regions, both in the forward and reverse potential scan for methanol oxidation on platinum. In both methanol oxidation and formic acid oxidation, a negative differential resistance (NDR) was observed in the potential regions that possess a negative d.c. polarization slope, and was attributed to the formation of surface oxide which inhibited the oxidation of methanol or formic acid. EIS spectra for formic acid clearly showed the presence of an additional low frequency relaxation at potentials where we expect adsorbed dissociated water or platinum oxide to be present, implying that more than one single coverage is required to describe the state of the surface and the kinetics. Two potential regions with hidden negative differential resistance (HNDR) behaviour were identified in the positive-going sweep, one prior to platinum oxide formation, assumed to involve adsorbed dissociated water, and one just negative of the main oxidation peak, assumed to involve platinum oxide. Oscillatory behaviour was found in the formic acid oxidation on platinum by adding a large external resistance to the working electrode circuit, which means that there is no longer true potentiostatic control at the interface. By revealing the system time constants, impedance measurements can be used to assist in explaining the origin of the oscillations. In the case of formic acid, these measurements showed that the oscillations do not arise from the chemical mechanism alone, but that the potential plays an essential role. Preparation and optimization of gas-diffusion electrodes for high temperature polymer electrolyte fuel cells based on phosphoric acid doped polybenzimidazole (PBI) membranes was performed. This fuel cell allows for operating temperatures up to 200 °C with increased tolerance towards catalytic poisons, typical carbon monoxide. In this work we employed pure hydrogen and oxygen as the fuel cell feeds, and determined the optimum morphology of the support layer, and subsequently optimized the catalytic layer with respect to platinum content in the Pt/C catalyst and PBI loading. A smooth and compact support layer with small crevices and large islands was found to be beneficial with our spraying technique in respect to adhesion to the carbon backing and to the catalyst layer. We found that a high platinum content catalyst gave a significantly thinner catalyst layer (decreased porosity) on both anode and cathode with superior performance. The PBI loading was found to be crucial for the performance of the electrodes, and a relatively high loading gave the best performing electrodes.

In the past more and more advanced and fine-tuned processes for steel production have resulted in increased demands for new and more costeffective ferroalloys used as constituents in the processes. Casting techniques and equipment are continually studied for potential improvements. In order to ensure a high and consistent quality in the alloys and the casting equipment, the heat transfer from the alloy to the mould during casting must be understood. Research on free metal flows is scarce and to remedy this a doctoral study at the Norwegian University of Science and Technology was initiated. The study was limited to the region around the impingement point of the metal jet, because this is the critical area for both heat and mass transfer. The flow develops radially, first as a thin film spreading evenly over the surface. At a certain point the thickness of the film increases suddenly - known as a hydraulic jump. Only steady-state conditions on a flat plate (without accumulation of fluid) are studied. The jump develops before the flow reaches the edge of the plate and maintains this position until steady-state conditions are obtained. This system is believed to be a good approximation for the initial conditions during the filling of an open mould. This is also the period when the thermal load on the mould is greatest. Numerous practical and mathematical simulations have been carried out and a relatively simple analytical model depicting the surface rofile of the liquid metal including heat transfer to the surroundings has been developed. The computational fluid dynamics code FLUENT was also used to compute the surface profile with the Volume-Of-Fluid technique, but with little success. The code was instead used to determine the flow and temperature fields inside the already established surface profile.Various laminar and turbulent flow models (variations of the k - εmodel) were tried and compared. Experiments with water were carried out for studying the flow field. Tin was used for heat transfer studies. Finally, these simulations were compared with results from the practical experiments. Introductory experiments were carried out with ferrosilicon with the intent to perform complete experiments with this metal.Measured heat flow usually exceeded predicted values, particularly in the stagnation region. Good agreement is shown between the results from the FLUENT simulations and the new analytic model, which shows good promise of acting as a useful alternative to the much more demanding numerical simulations.

An apparatus to measure compressive creep in carbon materials has been developed. Using the final experimental set-up five material properties could be measured in each electrolysis experiment. Creep, sodium expansion, compressive strength and E-modulus were measured for 3 commercial cathode materials at 25 and 980 °C with and without electrolysis. The sodium diffusion coefficient (D) was calculated from the sodium expansion results. Filler materials for cathode blocks, i.e., certain anthracite and petrol coke qualities, were exposed to sodium vapour to examine crack evolution. Creep The three commercial cathode materials were found to deform with time under compression at 25 °C, 980 °C and during electrolysis at 980 °C. Only samples from one block for each quality was studied, so care must be taken before extending the ranking to all classes of cathode materials. The ranking from low to high creep at 980 °C and during electrolysis was: Semigraphitized < Anthracitic < Semigraphitic The creep is larger during electrolysis than at 980 °C for all materials and the increase in creep from virgin to electrolysed material at 980 °C is largest for the anthracitic material. Repeated loadings did not influence the shape of the creep strain curve, which could be described by the expression (time)n. The magnitude of the creep strain ranged from 0.01 to 0.07 % with a load of 20 MPa held for one hour. The largest measured creep was approximately 0.35 % after 20 hours of electrolysis in the semigraphitic material. The stress-strain diagram of the anthracitic material is unchanged before and after electrolysis and exhibits a more linear behaviour than in the other materials. The stress-strain diagram of the semigraphitic and semigraphitized materials changes after loading and tend to increase after electrolysis. Cracks All anthracite grains cracked to some extent after being exposed to sodium vapour at 800 °C. The lowest heat treated grains cracked the most. Cracks through grains were also found in the commercial material during electrolysis. In the petrol cokes only grains calcined to 1500 °C with a structure characterised by a gradient from mosaic to flow was observed to crack after exposure to sodium vapour at 800 °C. Diffusion coefficient The diffusion coefficient of sodium in carbon during electrolysis has been calculated with three different solutions of Fick’s law and is found to increase with current density and graphitic character of the material. Two of the calculations were based on the expansion of the sample (penetration from bottom and radial penetration) and one on a rather few measured sodium concentrations. The diffusion coefficient was calculated to be in the range 8-10-5 to 5-10-4 cm2/s at current densities from 0.06 to 0.88 A/cm2, which is around 10 times larger than reported before (Table 2.1). The cryolite ratio did not influence D as the saturation time for samples in acidic or basic melt was the same. The ranking from larger to smaller diffusion coefficient in the studied materials was Semigraphitized > Semigraphitic > Anthracitic The diffusion coefficient increased with heat treatment temperature in some laboratory produced materials.

Tantalum diffusion barrier coatings were investigated as a way to improve the conversion efficiency of CIGS (copper indium gallium diselenide) solar cells.  Tantalum coatings were deposited upon silicon and stainless steel foil substrates using direct current magnetron sputtering (DcMS) and high power impulse magnetron sputtering (HiPIMS).  The coatings were characterized using scanning electron microscopy (SEM).  Cross-sectional scanning electron micrographs revealed that the HiPIMS coatings appeared denser than the DcMS coatings.