Thesis (M.S.) -- Middle East Technical University, 2005. / Keywords: e-commerce, mining sector, road map for e-commerce, strategy for e-commerce, steps of e-commerce, stages of e-commerce, application of e-commerce

Risers in offshore operations are subjected to corrosion during their service life cycle. The use of relatively inexpensive, high strength to weight ratio fibre reinforced polymer composite (FRPC) as a load bearing pipe repair sleeve is an emerging technology that is becoming common for offshore applications. Risers experience complex loading profiles and experimental investigations often incur substantial time, complicated instrumentation and setup costs. The main aim of this research is to develop a design tool for the repair of offshore riser that suffers from external corrosion damage on its surface using FRPC material. The simplest configuration of a fixed platform riser in the form of a vertical single-wall pipe is being considered. Characterization of the stress-strain behaviour of the FRPC laminate in the composite repair system subjected to various load profiles of a common riser is performed. The means of composite repair takes into account the ease of automated installation. The final repair method considers the use of unidirectional pre-impregnated (prepreg) FRPC that is assumed to be helically wounded around the riser. Finite element models of the composite repair system were developed via ABAQUS. Global analysis of the entire length of the riser was omitted as external corrosions usually occurs in a localised manner on the surface of the riser. Instead, local analyses were conducted where boundary conditions were applied to mimic an infinitely long cylindrical structure such as the riser. The local analyses FEA models were made to capture the stress-strain behaviour of the FRPC laminate subjected to different load profiles including static loadings such as internal pressure, tensile load and bending load. The design loads were calculated based on a limit analysis known as Double-Elastic Curve method developed by Alexander (2008). Proper element selection and mesh convergence were carried out to determine the FE model that can minimize the time and CPU memory needed for the simulation without compromising the accuracy of the results. The second part of this research integrated experimental tests to validate the FE model developed using the ABAQUS general purpose code. Due to constraints on cost and supply of materials and equipment, small-scale tests were conducted. Similitude relations were used to determine the scale properties between the model and the prototype. The final results showed that the FE model can represent the real-life tests of corroded riser repaired with off-axis FRPC laminate with great accuracy of more than 85%. Hence can be a useful tool for design and parametric study of the composite repair system. Using the validated FE model, an extensive parametric study of the composite repair system with respect to varying corrosion defects was conducted. The thickness and length of the repair laminate were compared to the ASME PCC-2 standard. Optimum thickness and length of the composite laminate were determined based on the maximum allowable strains computed using the Double-Elastic Curve method. In addition, varying fibre angle orientation of the unidirectional prepreg was considered as it is one of the main factors in helical winding. Based on the results from the parametric study, a simple relation was developed to predict the required thickness of the composite repair system subjected to combined loading. This relation combined with the developed FE model can be used to provide a quick design and performance validation of a composite repair system for offshore riser, which is the main novelty aspect of this research.

The 7175-T7351 aluminium alloy was studied to determine its suitability for the step-aside gearbox housing on the Rolls Royce Trent 1000 engine. The industrial motivation of this work was to reduce the weight of the gearbox housing using this lightweight material to ultimately improve the specific fuel consumption of the aircraft. This involved obtaining the mechanical properties of the aluminium alloy via a series of uniaxial mechanical tests with parameters based on the operating conditions of the gearbox housing during a typical flight cycle. Furthermore, a constitutive viscoplasticity model, with the inclusion of material ageing parameters, was developed to predict the material’s cyclic response under strain-controlled isothermal fatigue conditions at the gearbox housing’s operating temperatures. With this capability, a prediction for when the strength of the gearbox housing falls below the required design strength for safe use could be made. The room temperature hardness tests demonstrated the effect of time spent at elevated temperatures on the material’s hardness. It was found that the higher the soak temperature, the greater the initial rate of decrease in room temperature hardness and the lower the asymptotic value of hardness that was reached. For example, up to 24 hours of soaking at 200◦ C, the hardness decreased by 33%, and up to 1000 hours the hardness had decreased by 55%. For the same durations at 180◦ C, the hardness decrease was 17% and 47% respectively. Soaking at 120◦ C had an insignificant effect on the hardness of the material, indicating that the microstructure was thermally stable. Hardness testing could be used as a method to assess the strength of the gearbox housing for service monitoring during certifcation. Similar to the hardness tests, the elevated temperature tensile test results also revealed degradation in the mechanical strength of the alloy after prior soaking at elevated temperatures. The tests at 200◦ C on the as-received material decreased the yield stress by 31% and after soaking at test temperature for 20 hours prior to testing, the yield strength dropped by 52%. After a 2 hour temperature, the yield stress decreased from 220MPa to 165MPa which is alarming since the gearbox housing spends about 18 minutes at 200◦ C and 190MPa during climb. This suggests that in less than 6 flight cycles, the material’s strength will fall below the maximum operating stress of the gearbox housing and will be unsafe for continued use. Samples were soaked for up to 400 hours at 200◦ C and prepared for microstructural analysis. EBSD images showed that the grains were no significantly affected by the temperature exposure and showed no signs of coarsening. TEM and EDX analysis revealed that the majority of the particles within the grains were zinc-magnesium rich particles and were assumed to be MgZn 2 precipitates based on the TEM particle identification. The precipitate size and inter-particle spacing were found to increase with soak time. The change in monotonic yield strength was therefore attributed to the coarsening of these precipitates. The material characterisation suggested that, although the 7175-T7351 aluminium alloy initially appeared to have desirable mechanical properties, it is unsuitable for this or similar applications due to the rapid decrease in strength and thermally unstable microstructure. Furthermore, if an aluminium alloy is considered for this application, then it may be vital to account for material ageing behaviour. The unified, uniaxial viscoplasticity Chaboche model was implemented to predict the material response strain-controlled isothermal fatigue tests at 160◦ C and 200 ◦ C. A material ageing term was added to the model to account for the material ageing that decreased the yield strength with time. With this addition, two assumptions were made: 1) material ageing only affects isotropic hardening and 2) isotropic hardening can be de-coupled into material ageing (as a function of time at elevated temperature) and mechanical softening (a function of accumulated plastic strain). The tests at 160◦ C and 200◦ C showed that numerical and experimental results were in good agreement, providing accurate isothermal cyclic stress behaviour of the 7175-T7351 aluminium alloy. Furthermore, it was shown that the mechanical softening and material ageing components could be de-coupled. However, when the model was used to predict stress-controlled isothermal fatigue data and a cyclic stress relaxation tests, a number of deficiencies arose. The predicted ratcheting and ageing rate was greater than expected. The material ageing term may require an additional function to change the ageing rate depending on whether the material is elastically or plastically loaded. Norton’s creep power law could not predict term long stress relaxation behaviour but it was sufficient enough to describe to short- term viscous effects under the strain-controlled fatigue conditions. Despite these deficiencies, the model provided an initial point for a unified, viscoplasticity model for the 7175-T7351 alloy. Due to the rapid ageing of the material, the model could be used to predict if or when a material’s strength is unsuitable for safe operating use.

Laser sintering (LS), an additive manufacturing (AM) technology, allows for the production of 3-dimensional parts by fusing together successive layers of polymer powder without the need for tooling. Its potential and applicability, however, is still constrained due to the limited repertoire of materials available and the lack of detail in understanding both the important process-material interactions and consequently the requirements for the development of new materials. Past research has mainly focused on polyamide 12 (PA12) as the standard material, hence most of the empirically grown or theoretical, often idealized, process models are based on this polymer. As a result, it was shown that there are strong interactions between the material and the process leading to an undesired deviation of part properties. Thermoplastic elastomers (TPEs) for LS have gained more and more popularity for the production of, for example flexible parts in the recent past but they are a group of polymers that is neither well studied nor understood regarding their use in LS. Therefore, this PhD investigation has focused on TPEs in order to reveal their process specifics throughout the processing chain in LS. As the properties of parts manufactured by LS are, amongst others, influenced by the packing and flow efficiencies of the powders, the bulk (static) and flow (dynamic) characteristics of the observed TPEs were examined on the powder scale as well as their effects on the process and parts. The resulting part properties were evaluated in terms of their tensile properties, surface roughness and density. In contrast to previous studies which have rarely taken into account the characteristics of the un-sintered particles, this work provided a novel approach quantifying and describing the interconnection between the powder characteristics as well as its performance and the part properties, thus providing valuable input on future material design. As mentioned before, for typical semi-crystalline thermoplastics such as PA12, different idealized process models describing the Pre-Process specifics exist, but they might not necessarily be applicable for thermoplastic elastomers with significantly different thermal characteristics. Consequently, the important interactions during processing of TPEs have been studied by high speed and high resolution thermography helping to indicate the most important material properties in combination with calorimetric analysis. The resulting understanding of crystallization and melting behaviour helped to derive a design of experiments revealing the possible range for the process management in terms of temperature control as well as energy input and the resulting part properties. Moreover, it is known that polymers used in LS change their intrinsic properties due to processing conditions that are close to the crystalline melting temperature. As a result, within this PhD investigation the ageing behaviour of TPEs was studied. Both the powder and the sintered parts were examined for chemical and physical ageing effects. The results showed that the materials observed could be used without refreshing throughout the applied ageing cycles, however, changes in the processing behaviour as well as in the parts’ mechanical properties were evident. These changes were due to the differing ageing states of the LS-powder showing an increase in the particle size affecting the bulk materials packing density. In the literature, modifications in the rheological properties due to thermal loads during LS are already known on basis of PA12. It was shown that they tend to experience an increase in molecular weight with increasing processing cycles. In this work it was found that TPEs exhibit the exact opposite trend in a slight decrease of molecular weight likely to reduce the mechanical strength of tensile specimens. By using novel process adapted methods in order to reveal vital interactions of TPE powders for LS, this thesis makes a significant contribution to an AM database. In addition, the comprehensive observation and applications of these methods over the whole processing chain helped expanding the understanding of important requirements for the development of new polymers for LS and builds a substantial basis for future work and quality assurance.

A study was made of the rate of absorption of sulphur in grams per minute from sulphur dioxide - nitrogen gas mixtures by molten iron and iron containing four per cent carbon, The maximum rate of nitrogen flow and partial. pressure of sulphur dioxide was 350 ml/minute and 0·06 atmospheres respectively. The melts were contained in alundum and magnesia crucibles at 14000 - l550 [deg] C, the gas being blown on to the surface of the metal and occasionally bubbled through the metal for mixing purposes by means of an alumina tube. Samples or metal were taken periodically and analysed for carbon and sulphur by combustion methods.

A study on the chlorination of Basic Oxygen Steelmaking (BOS) by-product dusts followed by water leaching was conducted. The samples used for the studies sourced from the earth works and beneficiation of BOS byproduct from the stockpile. Comprehensive reports resulted from the excavation, beneficiation, physical and chemical analysis are reported. Zinc and iron removal studies from BOS by-product were investigated by means of chlorination roasting. Chlorination roasting of the BOS material by means of pyrometallurgical extraction was undertaking, assessing parameters such as roasting temperature, roasting time and chemical stoichiometry. A preliminary study was carried out using a solid chlorination agent, Ammonium Chloride (NH4Cl) in a muffle furnace for a 3 stage roasting conducted for 15 minutes to 180 minutes at 450C to 750C. Selected studies were carried out with 1-stage roasting. Further experimental studies then took place for chlorination using gaseous Cl2/N2 mixture gas in the tube furnace with 5 times chemical stoichiometry for a roasting time from 5 minutes to 90 minutes, with various temperatures from 150C to 750C. A 1 times and 2 times stoichiometry were added in selected procedure to foresee the effects of starvation agent. Additional surface area affects were also added to the chlorination with Cl2/N2 gas using bigger sample boat. Water leaching for 1 hour and 24 hours were conducted on the sample roasted with Cl2/N2 to investigate the leachability of Zn to assist further removal. The chemical and mineralogical composition of the BOS stockpile varied widely. The pH value were highly alkaline ranging from 10.2 – 13.5, while the moisture content showed a very wide range from 6 – 43%. The particle size analysis of the earth works excavation on the study site established five stockpile particle size fraction, ranging from as fine as <0.8mm to >38.22 mm. The composition of zinc ranging from 2.78% - 5.96%, while the iron content ranging from 41.36 to 62.18% respectively. This amount of iron has the potential for recovery and reused within a steel making process. It is possible to recover 97% of Zn and 30% Fe by roasting at 750C for a period of 135 minutes with NH4Cl salt. Roasting with 3 stages of NH4Cl addition proved to be more appropriate to increase the percentage of zinc removal. 97% Zn are removed Cl2/N2 gas for a period of 30 minutes roasting. The water leaching was conducted More than 90% Zn are recovered using bigger surface area sample boat with only 2.5 times stoichiometry. The extraction of Zn was greatly enhanced by the water leaching following the chlorination roasting using Cl2/N2 gas. 95% of Zn was leached while only 7% Fe was extracted after chlorination at 650C for 20 minutes roasting. It is possible to leach up to 98% Zinc after the chlorination roasting at 450C at 20 mins. The research has demonstrated that chlorination extraction has potential application in the steel sector for removal of Zn from steelmaking BOS dusts. In conducting the chlorination roasting, operational parameters such as temperature, roasting, chlorine agent stoichiometry, and surface area are important in determining the best operational condition. Hybrid process of chlorination roasting with leaching could highly assist in the further removal of Zn from BOS dust.

The growth in the use and wear of Ti-based alloy components in mining and offshore explorations has led to a search for techniques to re-engineer such components for reuse. The most desirable method of restoring/protecting the component surfaces is by hard-facing to enhance longevity in service. Laser cladding is one of the viable techniques to achieve a thick coating on such components which involves the addition of reinforcing particulates to improve surface properties such as hardness, wear and erosion resistance amongst others. A fundamental study and understanding of the resultant microstructure-property of the laser clad, hard-facing composite becomes necessary. In this study, laser cladding of Ti-6Al-4V wire with Spherotene particulate reinforcement and laser cladding of modified pre-blend of Ti-6Al-4V and TiB_2 powder were undertaken. The resulting physical and microstructural characteristics, hardness, and performance characteristics of laser clad composites were investigated. Samples from the as-deposited laser clad composites were characterised using optical microscopy, scanning electron microscopy (with chemical microanalysis) and X-ray diffraction. Performance characteristics were examined via erosion testing of the laser clad Ti-6Al-4V/Spherotene using plain and abrasive water jetting, and tensile testing of the laser clad Ti-6Al-4V/TiB_2 composite. The results showed that a crack and pore free clad containing as high as 76 wt.% Spherotene in the Ti matrix was achieved at an energy density of 150 J.mm^-2, 275 mm/min traverse speed, 700 mm/min wire feed rate and 30 g/min powder feed rate. The microstructure of the laser clad Ti-6Al-4V/Spherotene is characterised by nano-sized precipitates of reaction products (W and TiC) uniformly distributed in a beta-Ti solid solution matrix. Matrix hardness is enhanced by the presence of the reaction products in the Ti ranging between 410-620 kgf.mm^-2. Moreover, the modication made to the 90 wt.% Ti-6Al-4V/10 wt.% TiB2 feedstock by attaching the TiB2 to Ti-6Al-4V allowed uniform distribution of reinforcing element in the deposited composite. The composite microstructure on solidication is characterised by TiB eutectic needle-like features uniformly distributed in a Ti-rich primary phase. The hardness of the composite ranged between 440-480 gf.mm^-2. Tensile tests showed that the mean elastic modulus of Ti-6Al-4V/TiB_2 composite is 145 GPa, which is a 27% improvement when compared to that of Ti-6Al-4V. Erosion test indicated that the Ti-6Al-4V/Spherotene composite offered as high as 13 and 8 times resistance of that of Ti-6Al-4V when subjected to PWJ and AWJ impacts respectively.

This thesis covers in detail a study of the excavation of rock salt by roadheader, the factors affecting performance and finally a specification with operational results of a new production machine to suit the South African Coal Mining Industry. Dosco Overseas Engineering Ltd. the author's employer, is introduced. Reference is made to how, over the years, performance prediction has radically changed from a mere approximation to a position where an accurate value with a performance guarantee is a necessity. Reference is made to the Universities of Newcastle-upon-Tyne, Nottingham and Leeds who have been the main suppliers of rock testing facilities. The University of Newcastle-upon-Tyne has had further responsibilities for a specific test programme, funded by Dosco, to establish a Performance Prediction Methodology. A general introduction to the trial site at Domtar Salt, located in Canada, is given, along with the current mining methods and the particular aspects requiring consideration if machine mining were to be adopted. A detailed study over a twelve month period covering three main topics; fines production, performance rates, and cutter pick suitability is described. Results are discussed at length and valuable conclusions are drawn. Extrapolation of the results to predict the performance of a larger machine suitable to Domtar's high production requirements is shown. The ability to relate this study to other applications and, in particular, the aspect of pick penetration and its effect on machine design is discussed. A prediction curve suitable for South African coal is shown, along with the necessary calculations to enable a high production rate and the corresponding effect on machine design. Specification features, such as boom force, cutter head design and cutter motor power, are considered at length. The implications for the machine manufacturer for even larger, more powerful machines is shown. An early correlation· of findings is established by comparison to field results from a smaller single boom, Dosco roadheader. The study concludes that current or new machine design can be favourably influenced to reflect varying market requirements and that accurate prediction of machine performance is possible. A later, overall study of the TB2500 shows achieved production rates and comments in particular, on machine mining rate and available mining time. Recommendations having particular reference to the equipment suppliers involvement in the sales procedure are given.

Early produced water treatment technologies were developed before carbon dioxide emissions and hazardous waste discharge were recognised as operational priority. These technologies are deficient in the removal of dissolved hydrocarbons and dissolved heavy metal ions which have been identified as major contributors to the high environmental impact factor of produced water. The simultaneous removal of heavy metals and polycyclic aromatic hydrocarbon (PAH) from produced water via photoelectrochemical process was identified in this work as a produced water treatment alternative with the potential to virtually eliminate the cost for chemical reagents and high energy input. Several grades of simulated produced water were synthesised and used to understand different parameters necessary for developing a successful photoelectrochemical treatment. The process demonstrated in this work followed a simple two–electrode photoelectrochemical cell where heavy metals were recovered on a platinum electrode with simultaneous degradation of PAH (phenanthrene) on a photoanode, with the aid of sunlight (simulated) and an applied cell voltage of 1.0 V. Multiwall CNT-TiO2 synthesised via a modified sol-gel method served as the photoanode after it was immobilised at a loading of 2.5 mg/cm2 on a titanium plate. The results obtained from the photoelectrochemical treatment showed a recovery of 1.6 g/cm2 of lead, 0.2 g/cm2 of copper and 0.1 g/cm2 of nickel from produced water on a 0.1 cm diameter platinum electrode after 24 hours of irradiation with simulated sunlight at 1.0 V cell voltage and a simultaneous degradation of up to 16 % phenanthrene on the photoanode, which gives a potential of scaling up the process to a commercial throughput.

There is growing interest in Al-TM (transition metal)-RE (rare earth) amorphous alloys because of the combination of their good mechanical properties and corrosion resistance. However, the high critical cooling rate required to form the amorphous structure leads to difficulties in generating bulk amorphous material. Therefore, there has been greater interest in producing amorphous Al-TM-RE alloys as surface layers. In this study, wedge mould casting, laser surface melting (LSM) and large area electron beam (LAEB) surface melting were used to fabricate Al Co-Ce alloys in both crystalline and amorphous form. An eutectic Al 33Cu (wt.%) alloy was also used to quantify the solidification conditions based on the well-known relationship. The microstructures formed by different processes were investigated by scanning electron microscopy, energy dispersive X-ray spectroscopy, image analysis and X ray and glancing angle X-ray diffraction. In particular, LAEB surface melting was found to be able to provide a sufficiently high solidification velocity for the generation of an amorphous layer on the remelted surface of bulk crystalline Al Co-Ce alloys. Experimental results show that the LAEB treatment can remelt, homogenise the multiphase crystalline starting material and generate a predominantly amorphous layer, although it also caused cracking of the treated layer. However, the cracking was largely reduced in the laser refined starting microstructure. Laser microstructural refinement also improved the homogenisation and amorphisation generated by the subsequent LAEB treatment. The temperature field of multi-pulse LAEB irradiated Al-Co-Ce and Al-Cu alloys was numerically simulated through a finite difference method. The simulation results were generally consistent with the experimental results. The corrosion behaviour of Al-Co-Ce alloys with different microstructures was studied through potentiodynamic polarisation tests. Al-Co-Ce amorphous layer exhibited an enhanced corrosion resistance compared to the crystalline counterpart, although cracking in the amorphous layer greatly influenced the effectiveness of the amorphous layer protecting the substrate.