Effects of aggregate densification upon thickening of Kynchian suspensions

Abstract

A model is presented for the thickening of a raked suspension. The model is based on Kynch theory (Kynch, 1952), i.e. it describes systems with solids fractions sufficiently low that the solids have not gelled into a weight-bearing network. However the model incorporates a modification to describe how raking the suspension causes flocs or aggregates within it to densify. This floc densification opens up channels between the flocs through which liquid escapes, making the suspension easier to dewater. The densification theory presented here predicts profiles of varying solids fraction vs height in the settling zone in a thickener, information which is not normally available when designing thickeners via the conventional Kynch theory. Performance enhancements for thickeners due to raking can be readily determined, either in terms of increased underflow solids fraction or increased solids flux. As underflow solids fraction is increased, thickeners operated at a specified aggregate densification rate (or equivalently at a fixed settling zone height) tend to approach a 'fully densified' suspension state (defined as a point at which the extent of aggregate densification ceases to change with time), with improved thickening performance

Designing thickeners by matching hindered settling and gelled suspension zones in the presence of aggregate densification

“…Recently however a study has appeared (Grassia et al, 2014) suggesting how to incorporate aggregate densification into the framework of Kynch theory. This study would apply to situations where the degree of thickening required (measured by a target underflow solids fraction) is relatively modest so that either the underflow does not form a gel or else it is just barely gelled, with very significant amounts of ungelled material elsewhere in the thickener.…”

Mathematical modelling of batch sedimentation subject to slow aggregate densification

“…Previous work (Usher et al, 2009;van Deventer et al, 2011;Zhang et al, 2013a,b;Grassia et al, 2014) has employed the value D agg,∞ = 0.9 which is also adopted here. Since the interest here is ultimately in a pseudo-steady state model here, the value of A does not strictly speaking need to be specified: changing the value of A merely changes the rate at which the system evolves through a given sequence of pseudo-steady states.…”

“…There have been a number of recent studies in the literature attempting a mathematical description of aggregate densification (Usher et al, 2009;van Deventer et al, 2011), and its effects on suspension dewatering processes (Zhang et al, 2013a,b;Grassia et al, 2014). These introduced a so called densification rate parameter (van Deventer et al, 2011;Zhang et al, 2013a;Grassia et al, 2014), the reciprocal of this rate parameter giving a characteristic time for aggregates to densify.…”

“…These introduced a so called densification rate parameter (van Deventer et al, 2011;Zhang et al, 2013a;Grassia et al, 2014), the reciprocal of this rate parameter giving a characteristic time for aggregates to densify. There are, as a result, three relevant time scales applicable to solid-liquid separation in a sheared batch settler: a characteristic sedimentation time (i.e.…”

“…More information about the models and underlying assumptions can be found in Usher et al (2009);van Deventer et al (2011);Zhang et al (2013a,b); Grassia et al (2014).…”