A series of isothermal and cyclic oxidation tests were carried out on two thermal barrier coating systems consisting of a CMSX-4 nickel-based superalloy substrate, NiCoCrAlY overlay bond coat applied by two different techniques and a yttria stabilised zirconia topcoat applied by electron beam physical vapour deposition (EBPVD). The bond coats were applied by either high velocity oxy-fuel spraying (HVOF) or EBPVD. Isothermal oxidation tests were carried out at 950°C, 1050°C and 1150°C for both coating system for up to 3000 hours. Cyclic oxidation testing was conducted at 1150°C in one hour cycles to coating failure on both coatings. A detailed examination on the oxide thickness was conducted on all specimens, along with characterisation of the bond coat and TGO. This was coupled with examination of specimen cross-sections for cracking and signs of coating degradation. Isothermal oxidation showed sub parabolic oxide growth consistent with the literature. Detailed analysis of oxide thickness showed a normal distribution but with increasing standard deviation with increasing oxidation time. Both bond coats were dual phase, β + γ. The EBPVD applied bond coat only, was found to contain yttrium rich precipitates in the bond coat and TGO. Both coatings showed no increase in surface roughness after either isothermal or thermal cycle testing. Short sub-critical cracks were observed at the TGO/topcoat interface in the HVOF applied bond coat only associated with the flanks of asperities. Coating failure in both coatings occurred at the TGO/bond coat interface on cooling and was likely driven by the thermal expansion mismatch between the TGO, topcoat and substrate. The initial mechanism of crack formation was not determined conclusively but could be a wedge cracking type mechanism.

The work presented within this thesis describes an investigation into the selective flocculation behaviour of coal/shale slurries using commercially available polyacrylamide polymers. It is considered that selective flocculation may eventually offer a method of recovering the valuable combustibles which are presently discarded. To date, although numerous, mainly laboratory scale, investigations using organic polymers have been conducted on a variety of mineral combinations, little work has been undertaken on coal/shale slurries. Selective flocculation depends essentially on differences in the adsorption of a flocculant on different minerals, which in turn depends on the comparative mineral surface chemistry. The effective dispersion of one component is also essential i.e. it must form a relatively stable suspension, either inherently or after modification with a suitable reagent. The effect of these two factors on both single specie flocculation and selective flocculation is discussed in some detail. Testwork was conducted throughout using pure coal and shale samples in order to minimise any natural variation in surface properties which may occur. In order to define possible conditions under which a separation may be feasible, the flocculating ability of various types of polyacrylamides on separate coal and shale slurries was initially established. However, the predictions were not entirely successful with the results indicating that, in a 50/50 w/w mixed slurry, under approximately neutral conditions, coal was the preferentially flocculated component. This was apparent for polymers of various molecular weights and ionic characters. Variations in the surface chemistry of the coal and shale may be influenced by several factors such as pH, dispersant concentration and promoter effects. Solids concentration is another important physical aspect. Each of these were studied and discussed. Zeta potentials of the individual coal and shale samples at various pH levels were also determined and related to the selectivity and efficiency of separation under similar conditions. It was apparent that although coal was preferentially flocculated, the major problem was that of entrainment. This inevitably arises during the formation of the coal floes, trapping some unwanted, and essentially dispersed, shale. In addition polyacrylamide may possess a weak affinity for the shale surface. These two mechanisms lead to products with a minimum ash content of between 31 - 36%. It was also found that the floes formed were so weak that they ruptured even on gentle agitation - hence trapped material could not be released. The effect of pre-conditioning the coal surface prior to flocculant addition was also assessed and achieved some degree of success, but again products of below 30% ash were not possible. It was therefore concluded that while coal may be selectively flocculated from a coal/shale slurry, using commercial polyacrylamide polymers, without an additional cleaning stage(s) the system is not economically viable.

An investigation has been made into the behaviour of short fatigue cracks in the β-processed titanium alloy, Ti6Ss. The effect on the material of a number of different variables was studied, particularly microstructure, as well as stress level, specimen design and specimen size. Initiation of cracks was found to occur at similar microstructural features in smooth specimens, regardless of a-platelet morphology. Cracks initiated from slip bands across small bundles of α-platelets, despite the presence of longer, more intense slip bands across α-colonies. α-platelet morphology was found to have a significant effect on short crack propagation rates when crack length was of the order of, or less than, the α-colony size - the more aligned the α-platelets, the greater the short crack propagation rates at equivalent ΔKs. Increasing the prior β-grain size (and hence the a-colony size) led to higher short crack propagation rates at equivalent ΔKs. Specimen design appeared to have an effect on short crack growth rates a uniaxial tension specimen design gave slightly higher growth rates than a four point bend specimen design. Variation in specimen size had no effect on short crack growth rates for the same microstructural condition, but a larger specimen size was found to reduce the scatter in specimen lives observed in smaller specimens. Short crack growth rates were apparently insensitive to stress level when compared at the same nominal ΔK values. Crack shape was found to vary considerably at crack depths below approximately 0.75mm. At crack depths greater than this value, cracks took an approximately semi-circular form. In keeping with many recent studies of short crack behaviour, short cracks were found to grow faster than long cracks at the same nominal ΔKs and at ΔKs below the long crack threshold value, ΔKth· A case study was carried out to determine the behaviour of short cracks in a cheese forged from an ingot slice of IMI685. Crack growth behaviour was found to be the same as that of Ti65s, but initiation behaviour was affected by the presence of pores in the IMI68S, leading to a significant reduction in fatigue life.

An investigation has been made of the growth of An investigation has been made of the growth of short fatigue cracks in aluminium and titanium alloys, with special emphasis on the effects of microstructure. For the commercial aluminum alloys examined (7010 and 2014A) a degradation in short fatigue crack resistance accompanied an increase in the degree of ageing, as is commonly reported for long, through thickness cracks. The short cracks however, were seen to propagate at substantially faster rates than conventional, long cracks at the same apparent applied ΔK and R - ratio. The similar, faceted fracture produced for both types of crack suggests that the same propagation mechanism was operative, despite the differences in growth rate. Discontinuous growth was observed for microstructurally short cracks, associated with the crack tip being held up over large numbers of cycles at grain boundaries both at the surface and in the interior of the specimen. The discontinuous nature of propagation and the anomalously fast growth continued until the maximum plastic zone size ahead of the crack tip approximated to the grain size of the material. At greater crack depths propagation rates, at equivalent ΔKs, for long, through thickness cracks and short cracks showed reasonable agreement, small discrepancies being explained in terms of the variation in closure contribution for the two types of crack. Similar propagation characteristics were observed for the two titanium alloys studied (IMI 318 and IMI 550) with substantially faster propagation for microstructurally short cracks than for long cracks at equivalent ΔKs. In contrast to long crack behaviour however, an increase in grain size resulted in a deterioration in short crack propagation resistance. The microstructure which was found to be the most resistant to the growth of short cracks consisted of fine primary α and transformed β grains with the volume fraction of each type of α approximately equal. The grain boundaries between the two distinct α-morphologies were seen to be particularly effective in reducing the propagation rate. In contrast, a coarse, aligned, Widmanstätten α-morphology exhibited the worst resistance to short crack growth. Even though colony and prior β grain boundaries were useful in retarding crack growth, propagation across packets of Widmanstätten laths was very rapid. Crack shape effects were considered to be important. A study was made to assess the variation in crack shape and the effect on crack propagation, particularly in non-equiaxed microstructures. Cracks with depths approximating to the grain size were seen to vary significantly in shape, though they all eventually took a semi-elliptical form, with half surface length/ depth (a/c) ratios between 1.0 and 0.8, at greater crack depths. It is suggested that a better correlating parameter for short crack growth is crack area rather than surface crack length which is commonly used at present. In this way, the effects of crack shape can be accommodated.

Easily accessible oil reserves are currently decreasing, leading to an increase in more complex offshore deep-sea drilling programs, which require increasingly greater depths to be drilled. Such wells are commonly drilled using oil based muds, which leads to the production of drilled rock fragments, drill cuttings, which are contaminated with the base oil present in the mud. It is a legal requirement to reduce oil content to below 1 wt% in order to dispose of these drill cuttings in the North Sea and microwave processing is suggested as a feasible method of achieving the desired oil removal. However, there are currently gaps in our understanding of the mechanisms behind, and variables affecting, the microwave treatment of oil contaminated drill cuttings. The work described in this thesis seeks to address some of these gaps in knowledge. There were three main objectives for this thesis: (1) quantification, for the first time in the literature, of the main mechanisms driving oil and water removal during microwave processing of oil contaminated drill cuttings, (2) determination of key variables affecting performance during pilot scale continuous processing of oil contaminated drill cuttings and, for the first time, (3) treatment of drill cuttings with microwaves continuously at 896 MHz. Bench scale experiments carried out in a single mode applicator were used to quantify the mechanisms involved in oil and water removal from drill cuttings. It was found that both vaporisation and entrainment mechanisms play a role in oil and water removal. Vaporisation was the main mechanism of water and oil removal, and typically accounted for >80-90% of the water and oil removed. For oil removal, vaporisation of the oil phase accounted for 70-100% of the overall removal. The absolute amount of water entrained and vaporised was found to increase with increasing energy input and power density. However, as a percentage of the overall amount removed, entrainment was found to increase with increasing energy input. This was mainly due to higher heating rates at higher energy inputs, leading to pressurised, high velocity steam, which increased liquid carry-over (entrainment). Both the drill cuttings sample composition and applicator type were found to have an effect on the extent of entrainment/vaporisation. Samples consisting of a higher overall liquid content, tended to have a greater amount of surface liquid content. This led to a greater potential of carry over when steam generated internally left the sample. Increasing the power again led an increase in entrainment in this case. Different applicators were found to impact the electric field strength and power density within the water phase of the sample. Oil removal in multimode applicators progressed mainly through vaporisation (steam distillation) until the water content was sufficiently low to generate steam at a velocity high enough to entrain liquid droplets. When treatment was changed to single mode operation, entrainment occurred at an earlier stage, probably due to higher electric field strengths and power densities. It was also noted that the vaporisation mechanism of oil was more efficient at higher field strengths and powers, which could again be attributed to superheating and higher velocity steam, which enabled better mixing and heat transfer. Experiments were also run to determine the main variables affecting the performance of continuous processing of cuttings. Overall continuous processing showed a substantial improvement in the energy required, 150 kWh/t vs. >250 kWh/t, to reduce the oil content of a drill cuttings sample to 1 wt%. It was found that the initial water and oil content of the sample, as well as the sample particle size distribution, had the greatest effect on the efficiency of continuous processing. The effect of initial water and oil content on residual oil content was investigated methodically for the first time for continuous microwave processing of oil contaminated drill cuttings. An increase in initial oil content was found to have a significant impact on the energy input required to treat the sample to 1 wt% oil content. As the oil content increased, the energy input required increased exponentially, mainly as a result of the change in the physical structure of the sample. An increase in the water content led to an increase in energy input without any additional benefit to oil removal. However, as the water content was increased it was noticed that the theoretical energy input required to heat the entire sample approached the actual value measured for the energy input. This occurs as a result of the increasingly greater bulk dielectric properties of the sample as a result of higher levels of water content, which in turn leads to a higher efficiency in the conversion of microwave energy to heat in the sample. The effect of particle size on oil content distribution and removal was investigated. Oil content was found to be substantially higher in particles of size <1.0 mm, with removal also being significantly higher in this particle size range. However, as the majority of the samples tested, >80%, consisted of particles >1.0 mm, this improved removal is diluted by the performance of the coarser particles. The improved removal in finer particles is likely to be due to larger surface area, reduced path length within the particles and potentially higher electric field strength. Finally, samples processed continuously using a continuous microwave setup at 896 MHz showed improvements over both continuous microwave treatment at 2.45 GHz and bench scale setups. Increasing the f10wrate of the system at 896 MHz was also found to improve oil removal efficiency, which can be explained by the higher power requirements that would be required to maintain the energy inputs observed at the lower flowrate. Increasing the power leads to improved heating rates and thus increased removal rates through entrainment and vaporisation.

A problem of potential water occurrences exists, whenever coal reserves are worked beneath either surface or sub-surface water bodies. The object of this work is to examine hydrogeological techniques and problems which can be associated with longwall coal extraction. Current U.K. mine drainage problems are discussed along with the various means by which water can enter a working horizon. An area of principal concern is the potential access of water via fracture networks associated with subsidence profile formation. The role of subsidence development with respect to different geological conditions is considered. Similarly, parameters which control the potential yield of an aquifer, as well as the techniques available for assessing them, are also discussed, with particular reference to Coal Measures strata. A reappraisal of existing test site data, collected by Nottingham University Mining Department is undertaken and concludes that permeability changes can be linked to subsidence profile development. Permeability monitoring techniques can be used for strata control investigations and reference is made to monitoring the caving characteristics of a massive sandstone roof. Similarly, problems connected with thick sandstone horizons and potential weight bump conditions are also examined. The effects of subsidence profile formation on the undersea coalfield workings of North-East England are examined with respect to the NCB 1968 guidelines for working under the sea. A geological and hydrogeological reappraisal, concludes that water occurrences can be expected in any area where a potential aquifer horizon has been displaced by faulting into close proximity with a working horizon.

Autogenous laser beam welds were made in sheets of the aluminium alloys 8090, 8009 and 6061. The Al-Li based alloy 8090 was subjected to both continuous wave CO2 and pulsed Nd:YAG thermal cycles with average powers of 1.5-3.8 kW and 0.8- 0.9 kW respectively. The two techniques were compared for their influence on the 8090 solidified weld pool shape, the fusion zone microstructure and microhardness, the HAZ and the susceptibility of the fusion zone to post-weld heat treatment. It was found that CO2 keyhole welding is preferable to Nd:YAG welding, under the welding conditions investigated, as essential elements such as Li and Mg were lost by evaporation during Nd:YAG processing. Microscopy of the 8090 CO2 weld fusion zone revealed that the solidification mode was sensitive both to the temperature gradient and growth rate during solidification, with a transformation from cellular to equiaxed dendritic growth occurring from the weld pool edge to the weld centre. The secondary dendrite arm spacing was found to be 2-5µm and the metastable phase δ (Al3Li) was present after welding with a very fine homogeneous distribution of -5nm diameter spheres. Porosity was identified as a major welding defect and was attributed to two distinct formation mechanisms. Firstly, the release of hydrogen gas during welding caused spherical gas bubbles throughout the weld pool. Secondly, the delicate balance of forces within the keyhole resulted in larger irregular shaped pores at the weld centre-line towards the weld root. This second type of pore was virtually absent in full penetration welds. The influence of heating rate to the solution treatment temperature on 8090 CO2 weld metal was assessed and the weld metal grain size was found to be most uniformly small after a heating rate of 1 K/min. A TEM investigation confirmed that the grain boundary pinning dispersoid β(Al3Zr) was responsible for inhibiting grain growth. However, the microhardness and notch-tensile strength of the CO2 weld metal did not depend on the weld metal grain size. Instead, it was suggested that the residual cast structure was responsible for determining the mode of failure and fracture strength. Microstructural studies of CO2 laser welds in RSIPM 8009 and wrought 6061 confirmed the cooling rate of 102-103 K/s predicted for CO2 welds in 8090. However, the 8009 CO2 weld metal did not solidify by epitaxial growth from the fusion boundary, which was the case for CO2 welds in 8090 and 6061. Instead, solidification in 8009 weld pools occurred via many isolated events on primary intermetallic particles. The intermetallic particles had the stoichiometry Al4.5(Fe,V,Si) with the AI.mFe tetragonal lattice parameters. It was qualitatively shown that the 8009 weld microhardness had an inverse relationship with the volume fraction of intermetallic particles. Chemical analysis of the 6061 weld metal confirmed that even when aluminium is alloyed with volatile elements such as Mg it is mostly retained within the weld pool during CO2 keyhole laser welding. It was found that a much higher power was required to obtain a deep penetration weld in 6061 than in either of the other alloys.

Surface mine planning involves the results of algorithmic numerical calculations being used by engineers to make informed decisions relating to the design. The Department of Mining Engineering at the Unversity of Nottingham has in the past been involved in developing modular algorithmic packages. The emphasis of the computer research has now altered. Smaller specialised systems are now being developed to cover individual aspects of the design process. Artificial intelligence techniques are being introduced into the mining environment to solve the planning problems often associated with the large amounts of uncertain information needed by the engineer. This thesis is concerned with the development of MINDER, a decision support system capable of assisting the mine planner in the complex task of optimum surface mining equipment selection. An expert system shell has been used to create a series of individual application modules, each containing a multi-level knowledge base structure. An information handling system has been developed which is capable of storing consultation information and transfering it between knowledge bases and between application modules. Once an effective method of information handling had been achieved the flow of control between the system knowledge bases was rapid and followed complex inferencing routes. Most of the commercially available packages mathematically model a deposit, calculate volumes and simulate operations. One of the aims of the MINDER system was to integrate with other software, for example MINDER is capable of reading volumetric and material information from Surpac mine planning software. Geological data and manufacturer’s equipment specifications are stored in DbaseIV databases. The expert system is capable of writing macros based on the consultation and performing complex relation operations involved in the elimination and ranking of equipment. In a similar manner macros are written to control the simulation package GPSS which used to simulate operations using the selected equipment. A range of ‘in-house’ Pascal software is used for numerical calculations and matrix manipulation, an example of this is the fuzzy logic software used to handle uncertain information. Another aspect of the project is an investigation into the use of machine learning techniques in the field of equipment selection. Knowledge induction software has been used to induce new rules and check those produced in the MINDER system. Various experiments have been carried out using neural network software to produce equipment selection models. Training data taken from the mining industry was used on both these systems and the results were tested against MINDER consultation results.

Plain-fatigue and fretting-fatigue tests have been carried out on the two peak-aged, high-strength aluminium alloys, 7010 and 8090, in the unpeened and shot peened conditions. Shot peening produced a highly deformed surface containing many deep dimples and folds from which fatigue cracks rapidly initiated. Although initiation was early, propagation through the peened surface layers was retarded by the residual compressive stresses introduced by the peening treatment; this retardation being greatest at a depth corresponding to that of the peak residual compressive stress. Shot peening improved the plain-fatigue properties of the 7010 alloy at high applied stresses and this improvement decreased as the applied stress was reduced. A reversed trend was observed in the 8090 alloy with shot peening being beneficial to the fatigue properties at low applied stresses and detrimental at high applied stresses. The poor resistance of the 7010 alloy to fretting-fatigue was eliminated by shot peening and this was again attributed to the residual compressive stress induced by the peening treatment. Since the 8090 alloy in the unpeened condition is relatively insensitive to the effects of fretting, shot peening resulted in only a minor improvement in the fretting-fatigue behaviour of this alloy.

In recent years the drivage advance rates achieved within the UK coal industry have increased. In the 1980's average drivage rates were 35m per week compared to the 100/150m per week possible today. These extended rates of advance have resulted in an increase in the potential methane, dust and heat generation within the vicinity of the drivage face. In order to effectively disperse this additional pollutant load a controlled increase in air quantity is required. Although advance rates have changed, current auxiliary ventilation practice has not. UK mining law requires that the fresh air must be delivered to within 5m of the face. This has lead to the wide spread adoption of the use of overlap auxiliary systems within mechanised drivages, since a pure forcing system set at this distance from the face would lead to excessive airborne dust. UK mining law does not at present consider on-board mounted exhaust scrubber fans to constitute an effective overlap fan within mechanised drivages. Consequently an additional overlap exhaust fan is required to be installed within such drivages. In an attempt to determine whether working conditions could be safely and economically improved within mechanised rapid development drivages utilising an on-board mounted exhaust fan, a series of preliminary full scale gallery trials were conducted. A summary of the principal findings of these trials is presented together with an outline of a series of representative CFD simulations. This thesis examines the accuracy of CFD simulations for auxiliary ventilated headings. This is achieved by utilising Laser Doppler Anenometry (LDA) in a scale model representative of an underground heading and a detailed underground measurement programme conducted in production headings. These measured airflow values are then compared with representative CFD simulations and conclusions