Abstract
Bubble-particle interaction is a key phenomenon in many industrial applications, for example
in mineral froth flotation. Flotation systems are typically characterised by high void fraction
of dispersed phases and often multiple surface active compounds are present. The complexity
of bubble-particle interaction has lead researchers to develop simplified models for dilute
systems and typically physical and physico-chemical aspects are left out.
This work discusses a modelling framework for analysis of bubble-particle interaction in the
presence of soluble surfactants. The model includes full momentum coupling between gas,
liquid, and solid phases using a coupling between Computational Fluid Dynamics (CFD) and
the Discrete Element Method (DEM) named CFDEM. CFDEM is an open source modelling
framework where the CFD code OpenFOAM and the DEM code LIGGGHTS interact.
To accommodate topological changes of the bubble surface during break-up and coallescence
the Volume Of Fluid (VOF) method was used. Solid particles are tracked in a Lagrangian frame
of reference and experience forces due to collisions and the presence of the gas-liquid interface.
A comprehensive model has been developed where particle-interface forces are modelled as a
hyperbolic function of the gradient of the phase fraction. Particles can be captured within the
interfacial region and can detach from the bubble when the balance of forces so dictates. DLVO
and non-DLVO forces, as well as inertial forces, form part the total stress balance and
contribute to the momentum equation of all phases. Variable interfacial tension is taken into
account by implementation of a volumetric transport equation for soluble surfactant in the bulk
fluid and within the interfacial gas-liquid region. The method is fully mass conservative and
combines higher order physical momentum coupling with physico-chemical momentum. The
sub-models used need further study, but to the authors knowledge the model presented is the
first to couple all momenta in a comprehensive modelling framework for bubble-particle
interaction.
The main value of this work is that the computational framework is modular and easily
extensible to include more accurate sub-models. The Lagrangian particles are in fact dynamic
lists that can be populated by the properties appropriate to the system. These properties
accommodate further development and help to identify future research needs in the field of
flotation modelling.