The aim of this investigation is to characterise and refine the physical properties of glass fibre reinforced polyester polymer concrete. This material is currently being employed by AV Mouldings (Pty) Ltd. to manufacture manhole and drain components according to specifications existing for cast iron covers. No specification exists for polymer concrete. In particular it has been found that there is a large market for Type 2A replacement manhole covers and frames due to the current problem in South Africa of the cast iron versions being stolen and sold for scrap metal. It has been found that polymer concrete covers manufactured to replace stolen cast iron covers (in existing cast iron frames) fail occasionally in service. The investigation thus focuses on the characterisation of glass fibre reinforced polymer concrete and analysis of the current standards with a view to establishing a new South African Bureau of Standards (SABS) specification for polymer concrete manhole components. The main testing procedure involved flexural testing of beam specimens. Preliminary tests were carried out to measure strength, toughness, strain rate sensitivity, and the effect of different reinforcing materials. Accelerated degradation tests were then conducted to establish the materials resistance to UV radiation, acids, alkalis, and various solvents. Different resins were evaluated, and experiments were conducted using graded aggregates, in an attempt to reduce the number of voids in the material. Vibratory moulding techniques and postcuring methods were also evaluated. The viability of employing silane coupling agents in polymer concrete was investigated in detail towards the end of the research. Redesign of the Type 2A replacement cover was then undertaken.

This thesis examines the relationship between the microstructural and the mechanical properties of poly-(propylene-ethylene) bi-phasic copolymers. The copolymers investigated covered a comonomer content ranging between 4 and 23 percent ethylene. Nine grades were considered, with variables such as the melt flow index, the degree of crystallinity, the molecular weight distribution and the effect of a nucleating agent being examined. These copolymers have been characterised in order to gain a better understanding of the interrelationship between the morphological structure and their physical, mechanical, thermal and thermo-mechanical properties. The toughness of the copolymers can be enhanced at low temperatures by increasing the ethylene content, at the expense of a loss in stiffness. A study of the microstructure using the scanning electron microscope indicates that a good balance between these two properties can be achieved through a uniform size and spatial distribution of the ethylene-propylene rubber particles within the polypropylene homopolymer matrix. The transmission electron microscope shows the ethylene-propylene rubber (EPR) to be agglomerates of smaller particles, with some crystallinity within the EPR being evident.

Bibliography: pages 90-97. / The new concept of hydropower has been found to be technically feasible in South African gold mines. Chilled mine service water is piped from the surface to deep level stope; where the hydrostatic pressure provides power for stoping machinery. This water varies widely in composition and acidity. High concentrations of sulphate, chloride and nitrate are present. These ions are derived from the leaching of oxidised sulphides from the broken rock, the fissure water and the dissolution of blasting fumes. In order to minimise the deterioration of stoping machinery by corrosion and synergistic corrosive abrasive effects, a compromise between selecting a suitable corrosion resistant material and treating the mine service water to an acceptable level of corrosiveness is being sought.

Metal matrix composites consisting of 6061 and 2014 aluminium alloys, reinforced with 10%, 15% and 20% alumina particulates and a 6061 alloy reinforced with 20% SiC particulates, have been characterised in terms of their behaviour under various tribological conditions. In abrasive environments, the wear behaviour of each composite is dominated by their ability to resist indentation by hard particles. Abrasion against fine grit particles leads to a reduced load per abrasive particle and a correspondingly significant reduction in wear loss. Reciprocating sliding wear tests, conducted in an aqueous environment and against hardened steel counterfaces, displayed composite wear rates that were up to three orders of magnitude below those of their monolithic alloys. This is attributed to the increased resistance to surface shear provided by the reinforcing particulates themselves and their constraining effects on the matrix. The particulates become load bearing and protect the matrix by reducing the metal to counterface adhesive wear. However, the counterface wear increases due to the interaction with the hard reinforcements. Transmission electron microscopy of the worn composites reveal the formation of a transfer layer and subsurface dislocation structures which are similar to those found in metals subjected to low amplitude fatigue. In contrast to the results for abrasive and sliding wear, the composites show increasingly inferior cavitation and solid particle erosion resistances with increasing volume fractions of particulates. This depreciating effect was especially evident for particle erosion and can be related to the inability of metal matrix composites to accommodate the increments of strain which accompany erosive processes. The mechanisms responsible for the various performances have been studied by scanning electron microscopy, optical microscopy and transmission electron microscopy. An attempt is made to reconcile the steady state wear rates of the reinforced and unreinforced alloys with their observed wear modes, microstructures and bulk mechanical behaviour.

A nuclear grade 316L stainless steel and a 508-111 quenched and tempered pressure vessel steel were studied for their stress corrosion cracking susceptibility. Cylindrical tensile specimens were subjected to slow strain rate testing at 75°C in aerated, aqueous solutions (distiled water with 1000ppm Cl⁻ or SO₄ = ions in solution) in a range of corrosion potentials. The 316L has been examined for sensitization and stress corrosion resistance. This study has shown that the peak degree of sensitization attainable in this material is well within the limits considered as safe by the nuclear power industry. This material is not susceptible to environmentally assisted cracking as long as the potential is kept below the pitting potential for the material. A single instance of intergranular stress corrosion cracking was noted when this material was tested in 1000ppm Cl⁻ solution at 440mV (SHE). Two casts of 508-111 have been examined: 508-A has been tested in the as quenched condition as well as after two tempering heat treatments, while 508-B has been tested in the fully tempered condition only. The mechanical properties of the 508 type materials are strongly influenced by the heat treated condition and mildly influenced by the service environment. In the quenched condition anodic intergranular stress corrosion cracking is severe in the chloride solution and it is argued that the absence of intergranular cracking in the sulphate solution is due to the over aggressiveness of this environment. In all three heat treated conditions loss of ductility is more pronounced in sulphate solutions than in chlorides. Transgranular cleavage is evident in strongly cathodic conditions and this is ascribed to the ingress of hydrogen. The transgranular hydrogen embrittlement seems to be independant of heat treated condition. Rising load tests on fatigue precracked specimens have indicated that environmentally enhanced crack growth of existing defects does not occur for the conditions tested.

Bibliography: pages 91-96. / This study examines the effect that a partial substitution of nickel with nitrogen has on the deformation behaviour of a metastable austenitic stainless steel, AISI 301. The effect on the tensile deformation behaviour is studied in detail at various temperatures, and the effect on impact behaviour at room temperature is given brief attention. The uniform straining ability of a metastable austenitic stainless steel, such as AISI 301, which is used for stretch forming applications, is promoted by transformation-induced plasticity (1RIP), which depends on the manner in which deformation-induced martensite forms during straining. This includes both the rate of martensite formation, and the stage at which the martensite is formed. In particular, incipient necking is resisted when martensite forms gradually and selectively, preventing the formation and propagation of micronecks and microcracks. The microstructures of ten alloys, each having a type 301 base composition, but systematically varying nickel-nitrogen ratios, were characterized before and after tensile deformations using optical and electron microscopy as well as X-ray diffraction techniques. Tensile tests were performed on solution treated specimens at temperatures of 0, 20, 60 and 120°C. The martensite volume fraction present after a true tensile strain of 0.3 was measured, and the work- hardening behaviour of the alloys was characterized up to the point of maximum uniform elongation. All the alloys considered, showed fully austenitic microstructures at the solution treatment temperature (1050°C), and the indications are that nitrogen is fully dissolved. The austenite stability of the alloys however, varies at room temperature. Alloys containing approximately 5 wt% nickel, with a maximum nitrogen content of 0.28 wt%, contain up to 97% retained austenite, whereas alloys with 3.4 wt% nickel and the same maximum nitrogen content, contain only up to 63% retained austenite. The austenite stability is shown to affect the extent to which the TRIP behaviour occurs in the experimental alloys. In particular, the greatest ductility is provided by alloys containing approximately 5 wt% nickel and nitrogen contents in the range 0.14 - 0.16 wt%. This ductility is shown to be comparable to the AISI 301 alloy when deformed at 20, 60, and 120°C. The addition of nitrogen results in much increased strength compared to AISI 301, due to the interstitial solution hardening effect of nitrogen in austenite, and to a greater extent in martensite. Furthermore, pronounced serrated flow is identified in the experimental alloys when deformed at 60°C, whereas the behaviour is absent in AISI 301. Both these factors need to be considered when taking into account the formability of the experimental alloys. This study illustrates the potential for obtaining much cheaper 1RIP alloys relative to AISI 301, and at the same time indicates some of the limitations associated with alloying element substitution.

Bibliography: pages 125-130. / An investigation has been carried out on the effects of heat treatment on the microstructures and mechanical properties of a number of experimental 16 wt.% chromium dual-phase ferritic-martensitic stainless steels. A comparison was made between an alloy containing 2.5 wt.% nickel (low interstitial content [C + N = 0.03 wt%]), and three alloys possessing low nickel (0 - 1 wt.%) and high nitrogen contents (0.06 - 0.12 wt.%). Samples of AISI 304, 430 and 431 were included in the investigation for comparison with the experimental alloys. The microstructural response of the alloys to heat treatment was examined using light and scanning electron microscopy techniques. Tensile and Charpy V-notch impact tests were carried out on the alloys in their various heat treated conditions. Fracture surfaces, and deformation markings on the tensile gauge surface, were examined in the scanning electron microscope, while cross-sections of fracture surfaces were examined using light microscopy. Dilatometric traces were obtained for the experimental alloys in order to determine the effects of variations in composition on the inter-critical temperature range. The combination of good toughness and tensile strength that can be achieved in the low interstitial, nickel alloyed steel suggests that it could be a favourable alternative to both AISI 430 and 431 in many engineering applications. Toughness values superior to those of AISI 430 and 431 can be achieved in the high nitrogen stainless steels by tempering at 700°C, although the heat treatment results in a substantial loss in strength, and the low toughness exhibited by these alloys in the solution treated condition suggests that their weldability is no better than that of AISI 430 and 431. It is also shown that the formation of a lamellar ferrite/martensite compos.ite pha5e through intercritical annealing can provide attractive combinations of tensile strength, toughness and ductility in certain of the alloys. However, the ductility of alloys containing a lamellar composite phase is dependent on the o-ferrite content, and the toughness of the composite phase is adversely affected by a high nitrogen content. The yielding characteristics of ferritic-martensitic stainless steels are dependent on the hardness difference between the ferrite and martensite phases, and on the volume fraction of martensite. In addition, the morphology of the martensite phase exerts a strong influence on the ductility of dual phase steels. Microvoid initiation in the experimental alloys in the solution treated condition (1000°C/ lhour/air cool) occurs primarily by fracture within the martensite phase. In the 700°C tempered condition alloys having a high nitrogen content may be susceptible to intergranular fracture.

Bibliography: pages 62-65. / Glasses of the composition Na₂O-B₂O₃-X-Y and Na₂O-B₂O₃-Al₂o₃-X-Y were made where X and Y are two of the oxides CeO₂, HfO₂, ThO₂,Y₂O₃,ZrO₂ or Ga₂O₃. The glasses were either quenched or heat treated to promote phase separation and/or crystallization. The materials were subsequently leached in distilled water for periods of up to 96 hours. Leaching resulted in porous glasses or porous crystalline materials. The porous materials had high surface areas and good alkali resistance. A wide variety of compositions of these porous materials had BET surface areas between 100 and 413 m²/g. Selected porous glasses were chemically analysed. The heat resistances of two porous glasses were evaluated by observing the temperatures necessary for densification. Electron micrographs and X-ray diffraction scans were taken where necessary. Part of the glass forming regions for quenched samples of the system Na₂O-B₂O₃-Ga₂O₃-Y₂O₃ having Ga₂O₃ : Y₂O₃ ratios of either 3 : 1 or 3 : 2 were investigated. Part of the glass forming region for quenched samples of the Na₂O-B₂O₃-Y₂O₃ system was also investigated.

This thesis examines the cause and mechanism for the occurrence of parallel surface ridges during the deformation of two specific ductile metal alloys, namely AISI 430 ferritic stainless steel and an aluminium alloy designated AA3002. The investigation considers, in particular, the development of parallel ridges during uniaxial tension, and their effect on the overall surface roughening of the sheet material. A detailed account of the microstructure and texture of the individual sample sheet materials is presented and proposals for the mechanisms of surface roughening are based on plasticity analysis of the actual material data. Both the microstructural characterisation and the texture determination was carried out using mainly electron microscopy techniques. Electron backscattered diffraction techniques were used to measure the microtexture, and analysis of this data allowed the calculation of the plastic flow behaviour of discrete volumes of the sample material. The yield behaviour was implemented in a finite element model to simulate the material behaviour under uniaxial tensile conditions. Analysis of microtexture results has indicated that elongated texture clusters are visible in the aluminium sample material which exhibits severe surface roughening during elongation in the rolling direction. It is proposed that initially surface roughening is the result of a variation in plastic flow of the surface grains due to the local texture clustering. With continued straining, the condition described by the MK analysis for strain localisation is able to arise and this leads to through-thickness strain localisation and necking, and so results in the formation of a ribbed profile. Ridging in stainless steel on the other hand can be attributed to an asymmetric distribution of texture components or plastic flow properties about the mid-plane of the material. An asymmetric arrangement yield properties initiates the development of differential transverse strains about the mid-plane of the material. The variation in transverse strain in turn results in a series of localised bending events which, on a macroscopic level, produces longitudinal corrugations and an overall ridged surface morphology.

Understanding the creep deformation of high chromium steels in use in modern power plants has become important in predicting the behaviour and stability of these materials over their operational lifetime. At the deformation rates and conditions recorded in modern power plants, diffusional creep by vacancy migration is seen to be the dominant creep mechanism. However, the understanding of diffusional creep in particle stabilized materials is heavily incomplete. The aim of this project was to model the damage caused by diffusional creep, while considering the microstructure of high chromium steels and the evolution of this microstructure. This problem is addressed by expanding the existing Nabarro-Herring theory on lattice diffusion into a spatially resolved FEM model using MATLAB. This model focussed on adapting the Nabarro-Herring creep model to handle vacancy concentration changes over time. This allowed the model to produce the primary, secondary and tertiary creep stages present in experimental creep tests. As for microstructure, the focus was on adding precipitates (one of the strongest creep strengthening mechanisms) and voids (the largest cause of material damage). During creep exposure, precipitates were subject to coarsening while voids were subject to growth. The primary creep stage was formed by the initial rapid flux of vacancies into the body of the grain, due to large chemical potential gradients. A dynamic equilibrium of vacancy concentration would form within the grain, leading to the secondary creep stage. The creep rate produced was similar to that of the existing theory and it was found to decrease with the introduction of precipitates. This was evaluated by analysing the stress gradients caused by hard particles in a softer matrix. These stress fields lowered the stress in the grain boundaries and thus resulted in fewer vacancies being generated. Coarsening led to a reduction in the stress field distribution and thus resulted in creep strength loss in the material. The inclusion of voids was shown to decrease the initial creep rate, with void growth lessening this effect and leading to the tertiary creep stage. The initial strengthening was due to the void surface replacing the grain boundary as a source of vacancies. As the void surface is a very inefficient source, fewer vacancies were generated, resulting in lower diffusion rates. A slight steady increase in the creep rate over time was shown with the inclusion of void growth. The increase in vacancy generation was caused by the higher stress fields around voids. Initially the stress increase due to a loss in area was accounted for as a stress concentration around the void. Once this void grew too large in relation to the grain size, the stress concentration no longer accounted for all of the stress increase due to load bearing area loss. This resulted in the damage equation coming into play, causing a rapid increase in the stress throughout the grain and leading to the rapid tertiary creep stage.