Settling Dynamics of Suspensions in Viscoelastic Shear-Thinning Fluids Under Static and Sheared Conditions

Conventionally, miners utilise turbulent flow to operate slurry pipelines that move particulates across long distances. The advantage of this energy and water-intensive transport mode is its ability to use the turbulence eddies generated during flow to keep the flowing solid particles suspended in a horizontal pipe during a journey that could last for tens of kilometres. However, this advantage begs the question as to whether less water (and possibly energy) may be used to achieve the same transportation objective. This possibility, also known as laminar hydrotransport wherein particles are freer to settle during flow, has been scarcely explored due to the unresolved complexity in characterising how settling rates are affected by a host of parameters that arise from the rheology of the carrier fluid, solids concentrations and shear/flow. This lingering constraint in understanding inspired this research work which aimed to obtain empirically measurable data and use them to model settling rates of particulates in viscoelastic shear thinning fluids. To gain a better understanding of this problem, this PhD study conducted empirical investigations into how settling rates evolve within suspensions in viscoelastic fluids. These investigations were performed in analogous batch settling experiments- first under static conditions, then under sheared conditions. The working fluid used was a simulant made of aqueous Xanthan gum solution and glass microparticles; the former representing the viscoelastic ‘carrier’ fluid that often present in concentrated mining slurries, and the latter accounting for particles large enough to settle during flow. The measuring instrument used to acquire settling velocity and solids concentration data was an Electrical Resistance Tomography (ERT) unit, while a HR3 hybrid rheometer was used to measure the rheology of the carrier fluid. The settling mixture used in this study employed carrier fluids with the following rheology parameter ranges: flow consistency index, K = 0.29 - 0.74 Pa.s^n ; flow behaviour index, n = 0.28 - 0.38; and relaxation time, λ = 3.1 - 26.5s; and settling species whose sizes were between 126-533µm and concentrations in the carrier fluid were less than 5 vol%. Key findings in this work include a methodology to measure settling velocities from tomograms which, by default, are only capable of measuring local solids concentrations. Secondly, the relevance of the novel dimensionless ratios- the particle-acceleration number- in quantifying time dependent settling velocities were deduced. Also, settling was found to occur in three phases in both the static and sheared batch settling contexts: particle acceleration, particle deceleration and Stokes settling. Moreover, a fixed mathematical relationship was discerned between the acceleration and deceleration phase times in both the static and sheared settling contexts. Finally, the concurrent but independent effects of shear thinning, elasticity, solids concentration and shear rates on settling rates were characterised for the first time with a scalable model/equation. However, in both the static and sheared contexts, the correlation developed was limited to characterizing settling dynamics during the acceleration phase accuracy. Nevertheless, the accuracies of the model were acceptably high as they stood between 82-98%. Therefore, the model was deemed useful for analogous batch setting processes such as sedimentation tanks; and with a few assumptions, could be used to approximate the onset of a stationary bed during laminar pipe flow. Primary challenges overcome in this work include long experimentation times due to low settling rates of the chosen particle sizes, elimination of air bubbles during mixing within viscoelastic fluids, building a novel sheared settling setup consisting of a rotary belt constructed from LEGO bricks, and converting the novel settling velocity model to an appropriate format suitable for implementation in a Microsoft Excel spreadsheet. Future works would be geared towards improving the accuracy of the model and increasing its range of applicability. Future experiments would include measures to improve the model’s accuracy, such as: the use of perfectly monodisperse particles, using a novel rig that can investigate shear rates at various incidence angles on the settling particles, and conducting a full 3-D modelling of pipe flow while using pilot scale pipeloop experiment dataset to validate the model’s performance.