Abstract In order to improve the optimisation of mineral processing operations the rheological properties of slurries must be determined as accurately as possible under the conditions that closely resemble actual site conditions. The rheology of particles suspended in Newtonian fluids is well documented. However, the rheology of particles in non-Newtonian fluids has not been the subject of much investigation till now. The work conducted here attempts to fill this gap in knowledge. The rheological properties of slurries are heavily dependent on the solids concentrations and particle-solid interaction. At low solids concentrations, constant viscosity and Newtonian behaviour is observed, but as solids concentration increases the rheological behaviour becomes increasingly complex and non-Newtonian with viscosity becoming dependent on the shear rate. The nature of the non-Newtonian behaviour depends on the solid concentration, particle shape, particle size, particle size distribution and the suspending liquid rheological properties. The suspension/slurry may develop a yield stress and become time dependent in nature as structures develop within the fluid at higher solids concentrations. This study however, is primarily focused on the measurement of the rheological properties, where it is assumed that the fluid will be fully sheared and that the rheological properties will be unlikely to change with time. Shear thickening behaviour of slurries was the focus of this work. The aim was to investigate the slurry concentration region where shear thickening occurs. The first objective of the project was to develop a fluid analogue which will have similar rheological behaviour to that of concentrated tailings from gold mines so that it can be used as a test material to simulate the flow behaviour of the tailings in a pipe. The second objective of this project was to enable the prediction of flow behaviour in the pipe loop under certain conditions using the fluid analogue for slurry from Sunrise dam. In order to achieve the objectives, experiments were carried out to obtain a fluid analogue of a shear thickening slurry. CSL 500 and SR 200 rheometers were used for the characterisation of different fluid analogues and shear thickening mineral slurries. Malvern Sizer, model: mastersizerX v1.1, was used to obtain particle size distributions. A mini pipe loop system, located in the laboratory of the Rheology and Materials Processing Centre (RMPC) was used to get pipe line flow data for comparison with the rheometer data. A few fluid analogues with different suspending medium and different concentrations of glass spheres was tested before finally using, 48 vol% glass spheres in 1.8 wt% CMC solution as a fluid analogue for the mineral tailings obtained from Sunrise dam, WA. For comparison between the pipe line and rheometer data, all pipe line data (in the form of 8V/D) were converted to rheometer data (in the form of du/dr) using the Robinowitsch-Mooney equation. The above comparison indicated that it is possible to produce fluid analogue to simulate the flow behaviour of Sunrise dam slurry using a shear thinning suspending medium with high concentration of glass particles. Shear thickening flow behaviour was clearly observed in the rheometer while it was less predominant in a pipe line flow.

Transport properties of concentrated suspensions are of interest to many industries. Mineral slurries at higher solids concentrations have shown some rheologically interesting characteristics such as shear thickening, the increase of viscosity of a multi-phase mixture with increasing shear rate. The general literature on the rheology of suspensions records the presence of yield stresses, shear thinning and normal stress differences. Little is said specifically about shear thickening behaviour except for colloidal suspensions. The aim of this study is to examine the behaviour of coarse shear thickening suspensions and determine the causes of this phenomenon. The study intended to achieve the following objectives to; develop the appropriate techniques for rheometric studies of shear thickening suspensions; investigate the nature of particle-fluid interaction; develop a model of shear thickening behaviour as it occurs in non-colloidal suspensions and to develop a method of applying the rheology results to flows and flow geometries of practical relevance. The effects of wall slip dominate much of the literature of shear thickening materials. To investigate this aspect a significant portion of the experimental work examined the effect of shear thickening on torsional flow. The rheogram produced from parallel plate rheometry was reassessed as a non-controlled flow and a rheology model dependant analysis demonstrated that the effects of slip are considerably more problematic for shear thickening suspensions, particularly as wall slip is an increasing function of shear stress. As a consequence of the rheometric method described above it was observed that the rate of change of the first normal stress difference, N1, with shear rate changes as shear thickening commences for non-colloidal suspensions. N1 is initially negative and is increasingly negative at low shear rates. Additional rheometric analysis examined the transient effects in the behaviour of a non-colloidal shear thickening suspension. By employing large angle oscillating strain tests the strain required to initiate a shear thickening response was determined. Coherent back scattering of laser light experiments were able to show the change in orientation of the particles with respect to its rotation around the vorticity axis. After a viscosity minimum was reached the orientation became more random as particle rotation and lamina disruption occurred. This was considered to be the cause of the measured shear thickening. A model of shear thickening in concentrated, non-colloidal suspensions of non-spherical particles was developed. Based on hydrodynamic interaction in the Stokes flow regime, the flow of interstitial fluid subjected the adjacent particles to lubricating and Couette type forces, acting as a couple. When a series of force balances on a particle contained between two moving laminae are conducted as a time sequence, the particle orientation and motion can be observed. The model has qualitative agreement with several aspects of the experimentally observed behaviour of shear thickening suspensions, such as viscosity change with shear rate and concentration, and the first normal stress difference increasing with shear rate. Pipe line flow experiments were conducted on the model suspension. Particle settling produces unusual patterns in shear thickening suspensions, with an annulus of delayed settling near the wall.