A dissertation submitted to the Faculty of Engineering and the Built Environment, University of the Witwatersrand, in fulfilment of the requirements for the degree of Master of Science in Engineering. Johannesburg, 2016 / The feasibility of extracting iron from iron(III) oxide bearing materials with acetylacetone has been under investigation for many years. This is an alternate, environmentally friendly process for the recovery of iron compared to conventional processes that are energy intensive, have numerous costly process steps and produce large quantities of greenhouse gases. Iron(III) oxide bearing waste materials can be used in this process which reduces its environmental impact as it would not require waste storage. This study investigated the feasibility of reducing the reaction time of the liquid phase extraction of iron from iron ore fines by performing the extraction at elevated pressures and temperatures. It was found that that the extraction under pressure was dependent on temperature, pressure, particle size and solid to liquid ratio. It was found that at high temperatures and long extraction times, an unknown secondary reaction occurs that consumes the desired product, iron(III) acetylacetonate, and inhibits the recovery of these crystals. This results in lower extraction yields. It was found that the side reaction was largely dependent on the temperature of the system and the amount of iron(III) acetylacetonate present. The effects of the side reaction could be limited by lower operating temperatures and reducing the total reaction times. An optimum conversion of iron(III) oxide to iron(III) acetylacetonate of 47.2% was achieved for synthetic iron (III) oxide (> 95 wt% Fe2O3) at a total extraction time of 4 h, 160 °C, 0.025 g:1 mL, operating pressure of 1700 kPa, initial N2 feed pressure of 1010 kPa and 375 rpm stirrer speed. The optimum extraction of iron from iron ore fines (> 93 wt% Fe2O3) to iron(III) acetylacetonate was found to be 20.7% at 4 h, 180 °C, 0.025 g:1 mL and operating pressure of 1900 kPa, initial N2 feed pressure of 1010 kPa and 375 rpm stirrer speed. These are the optimum conditions where the side reaction is limited to improve the recovery and desired reaction conversion capabilities of the process. The operation under pressure yielded lower conversions than that of the atmospheric leaching process developed by Tshofu (acetylacetone water system under reflux). It was also found that it was not possible to reduce the extraction time and achieve comparable extractions when operating at higher temperatures and pressures. The formation of an additional unwanted product would also lead to unnecessary treatment costs in an industrial process. Hence, it was found that pressure leaching as an alternative is not currently viable due to the lower yields and associated high costs. Atmospheric leaching seems to be the most economically feasible option until a better alternative is found. / MT2017

No description available.

No description available.

A variety of iron precipitates formed in the electrolytic zinc industry were studied to estimate the possibility of magnetic filtration to augment conventional thickening/filtering systems. The precipitates included K,NH(,4),Na jarosites, (alpha) and (beta) goethite, (alpha) hematite and an industrially produced NH(,4) jarosite. The wet Frantz Isodynamic Separator was used to generate magnetic susceptibility data. Susceptibilities ranged from (kappa) = 6.9 x 10('-5) (--->) 15.5 x 10('-5) emu/cm('3)0e. Particle size of the precipitates ranged from 25 (mu)m to < 1 (mu)m. Magnetic filtration breakthrough curves were produced using a high gradient magnetic filtration technique. The role of fluid viscosity and velocity as well as magnetic field was studied. / All the precipitates proved filterable. A comparison was made of a physical and an empirical model of the breakthrough curves in the sizing of the magnetic filters required to treat thickener overflow at the CEZ Valleyfield plant. / For a volumetric flow of 100m('3)/hr electrolyte, containing 15g/L NH(,4) jarosite, 2 magnets of 2.6m diameter and 0.35m bed depth are required to lower solids contents to the present operating level of 3g/L.

No description available.

No description available.

No description available.

Abstract is not available

The raceway in the ironmaking blast furnace is a void in front of the hot blast tuyeres where coke and supplementary fuel burn to supply heat to the ironmaking process. The air velocity through the tuyeres is high (around 200 m/s) and this causes the coke particles to circulate in a rotating flow field inside the raceway. The size and shape of the raceway determine the gas flow dstribution, the reactions that occur, and the temperature profiles in the lower part of the blast furnace. It is for these reasons that the raceway has been extensively researched in the past. . Literature review revealed that forming and breaking of the raceway has not been yet studied. So, in the present study, we have concentrated our effort to study the formation and breaking of the raceway for different blast and bed parameters. The experiments have been performed in two-dimensional glass models. An attempt has been made to develop a correlation for each case. The formation of the void has been studied with change in parameters like density and size of the particles, bed height. The theory of the void formation has been derived from the fundamental principles. The formation of raceway, has been studied with the change in blast parameters like particle density and diameter, bed height and model width. It was found that when raceway is formed there is a sudden increment in pressure drop. The condition for breaking of the raceway, has been studied with change in blast and bed parameters like particle size and density, bed height and model width. It was observed that during the breaking of raceway, pressure decreases continuously. New semi-empirical correlations have been developed using dimensional analysis for formation of void, formation of raceway, and breaking of the raceway. The raceway growth also characterized with change in model width, flow rate, particle diameter, density of the particles and bed height. A few experimental results have been compared with published data. New semi-empirical correlation have been developed using dimensional analysis for the growth of raceway. Velocity of the gas exiting from top of the bed has been measured with the help of hot wire anemometer. It was observed that the velocity leaving from the bed is more on top of the raceway compared to the velocity leaving from the other parts of the bed. High velocity was observed near the wall of the model. Coefficient of wall-friction and angle of internal friction in presence and absence of gas were studied to explain the effect of bed height on formation of the void and raceway. It was observed that in presence of gas, coefficient of friction between the particles and wall and angle of internal friction between particles decreased, compared to the values in absence of gas. To study the effect of mixed particles, on the formation and breaking of the raceway, different sized particles in fixed proportion were taken. It was found that the experiments were more reproducible in case of uniform sized particles compared to mixed particles. It was also observed that for the same average particle diameter, mixture particles requires more velocity to form the void and raceway. New semi-empirical correlations have been developed using dimensional analysis for the formation of void, formation of raceway, growth of the raceway, and breaking of the raceway. Finally, an attempt has been made to quantify the various forces (pressure, bed weight & frictional forces) present in the raceway. Results show that further investigation is required in quantifying these forces properly.

Seven iron ore pellets, two sinters, and one lump ore were studied in CO-CO₂-H₂-N₂ atmospheres from 350°C to 750°C, simulating the upper stack of the ironmaking blast furnace. Experiments were performed in a flowing gas reactor on single specimens of each type of substrate. Two different measurements were made: (1) the carbon deposition and concurrent iron oxide reduction rate at 550°C in 30%CO, 10%CO₂, 2%H₂, and 58%N₂; and (2) the amount of carbon deposited during a programmed increase in temperature and change in CO-CO₂ ratio simulating the descent of an ore specimen in the blast furnace stack. The rates of the concurrent reaction were determined from mass balances based on gas chromatographic analyses of the CO, CO₂H₂, and N₂ in both the inlet and outlet gases and the continuously recorded mass of the specimen. The materials were examined as to chemical composition, internal structure, porosity, and surface area. Elemental analyses of single iron oxide grains were made by electron microprobe. Slag materials and composition, and crystallinity were determined by microprobe and X-ray diffraction. The results of the experiments show that carbon deposition occurs only in the presence of metallic iron which is produced from the concurrent reduction of Fe₃O₄. The degree of reduction is controlled largely by the structure of the substrate, but the carbon deposition is controlled only by the chemical composition of the substrate--specifically, silicon in the iron and the CaO to MgO ratio. In the blast furnace simulation, the carbon deposition increases for pellets fluxed with dolomite to a maximum with lime-fluxed pellets. The effects of H₂ and CO₂ on the reactions were investigated in the isothermal experiments using an Empire pellet. The CO₂ controlled only the reduction, and this by diffusion of the CO₂. The hydrogen in very small amounts enhanced the deposition of carbon, probably by eliminating the presence of the inactive iron carbides. Under blast furnace conditions, the changes in the operation when the chemistry of the ore feed is changed to fluxed pellets will be due more to the shifts in the available heat within the stack from carbon deposition than to the low temperature reduction of the ores, which does not change with the addition of the flux materials.