Chemical and Biomolecular Engineering - Theses
Permanent URI for this collection
Search Results
1 - 2 of 2
-
ItemSplashing due to gas injection( 1993)The splash event produced by a falling drop impinging on a bath of liquid was studied to enable a more fundamental understanding of the mechanisms of splash formation. An experimental method involving the use of a high speed photographic technique was developed to enable discrete splash events to be recorded at different time intervals. The effect of varying impact velocity, liquid viscosity and surface tension on the splash event was studied. The aspects of the splash mechanism that pertained to the projection of liquid above the liquid surface were monitored as a function of the experimental variables. Such aspects included the cavity depth, the Rayleigh jet behaviour and the extruded crown behaviour. Depending on the impact velocity of the falling drop, different shaped Rayleigh jets formed which varied in their manner of break up. Such changes in the Rayleigh jet behaviour could be related to the curvature of the interfacial cavity produced by the impact of the falling drop. As the impact velocity increased, the maximum height of the extruded crown increased. If the impact velocity exceeded a critical value, break up of the crown occurred. Increasing liquid viscosity had the effect of dissipating the energy of the system and hence producing less splash. Such viscous dissipating effects were quantified by the use of an energy balance. For liquids of a lower surface tension, there was evidence to suggest that a greater proportion of the incoming energy to the splash event was transferred to the extruded crown rather than to the cavity. Consequently, the splash produced from the break up and extrusion of the crown became more significant. A knowledge of the size of the bubbles reaching the liquid surface would assist in predicting the splashing behaviour of an industrial process which utilised gas injection.
-
ItemSlag-cleaning reactions in the KNS flash smelting furnace( 1991)The increasing cost of mining and concentrate production, together with the shift from traditional reverberatory smelting furnaces to more energy efficient smelting processes such as flash smelting, has made the slag cleaning step an indispensable part of current smelting technology. At the Kalgoorlie Nickel Smelter (KNS), slag cleaning is performed in an "appendage" to the flash smelting furnace. The aim of this project was to study the slag-cleaning reactions through plant measurements, thermodynamic modelling and laboratory experiments. Measurements taken at KNS showed that the slag and matte in the furnace were not in equilibrium. This was also predicted by the thermodynamic model and subsequently confirmed by laboratory experiments. The slag was over-oxidised relative to the matte and during equilibration iron in matte acted as a reductant for nickel oxide and magnetite in the slag. Thermodynamic modelling of the carbon-free slag-matte system indicated that the equilibrium concentration of nickel and magnetite in slag matte was about 0.4 and 7.5%, respectively. This is in good agreement with the results of the laboratory experiments in which samples of slag and matte from the smelting section of the furnace were allowed to come to equilibrium. The thermodynamic model also showed that further reduction of nickel from slag required the addition of carbon and that excessive reduction of slag would result in the reduction of FeO from slag and led to a decrease in matte grade. The latter was confirmed in a laboratory experiment performed in a graphite crucible. There was no difference between the performance of the various types of carbonaceous reductants at low N2 injection rates (0.2 to 0.4 l/min) in the laboratory experin1ents. In these experiments, the reduction rate of nickel and magnetite was relatively slow. When the experiments were repeated at a higher N2 injection rate of 2.0 l/min, the nickel and magnetite reduction rate increased significantly and it was found that the more reactive forms of carbon performed better than the less reactive carbon. The results suggest that with gentle stirring, reduction is controlled by mass transfer but with more vigorous stirring the reactivity of carbon is in1portant. The results of the laboratory experiments also found that the rate of reduction nickel and magnetite increased at higher temperatures. Injection of powdered carbonaceous reductants into slag in the laboratory experiments was more effective than lump addition of reductants in reducing nickel and magnetite from slag. Injection of reductants also resulted in greater utilisation of the volatiles in the reductants. The results suggest that the current practice at KNS could be made more efficient by lowering the nickel and magnetite concentration in slag before it enters the slag cleaning section. This could be achieved by injecting N2 to promote mixing and increase the rate of attainment of equilibrium between the matte and slag and/or by injection of powdered reductant, before the slag enters the appendage.