Investigation of factors effecting yield stress determinations using the slump test.

Abstract

Certain non-Newtonian fluids exhibit a yield stress which can be measured with a variety of instruments varying from very sophisticated rotary and tube viscometers to hand-held slump cones and cylinders of various sizes. Accurate yield stress measurement is significant for process design and disposal operations for thickenend tailings. The slump value was first related to the yield stress by Murata (1984). Later, that work was corrected by Christensen (1991) for an error in the mathematical analysis. Slump, based on a circular cylindrical geometry was first investigated by Chandler (1986). These concepts led to the study by Pashias et al., (1996) that formed the basis for the current research. The Flow Process Research Centre (FPRC) at the Cape Peninsula University of Technology developed a slump meter designed to lift the cone or cylinder vertically at controlled lifting speeds. In addition the simple hand-held cylinder which is an adaptation of slump cones which were originally developed by the concrete industry to determine the flowability of fresh concrete was also used. The vane technique was used as a control. Cones and cylinders made of stainless steel and PVC were fitted to the slump meter. The yield stresses of four non-Newtonian fluids at different concentrations were tested in four different configurations at different lift speeds to ascertain whether the measuring position, lift speed, slip, geometry, wall surface material, and stability has an effect on the value of yield stress measured. The effect of different predictive models was also ascertained.The cylinder, lump and cone models relating slump to yield stress was used in the dimensional analysis of the results. The objective of this work was to determine if the slump tests (cone, cylinder and the hand-held cylinder) would generate yield stress values comparable to those found using the vane technique. It was establised that there was no significant effect of lift speed, stability, geometry and wall surface material on the value of yield stress. The effect of measuring position on the value of yield stress calculated gave a difference of 25%. Using dimensional analysis, the lump model (Hallbom, 2005) more accurately predicts the material yield stress when using the hand-held cylinder as well as all the cone results (due to its specific geometry), and cylinder configurations, thus affirming the work of Clayton et al., 2003. It is concluded that, although the materials and concentrations tested induced errors within 40%, the hand-held cylinder shows promise as a reliable, quick and simple way of measuring the yield stress.