(219a) Process Intensification in Liquid-Liquid Extraction

The design and scale-up of extraction columns (i.e. columns with rotating internals) is still based on empirical rules and practical experience. Implementation of Computational Fluid Dynamics (CFD) into apparatus design is a promising engineering challenge due to the constant increase of calculation power. So far, detailed CFD-simulations of all interconnected phenomena (i.e. turbulence, bubble break-up & coalescence, swarm effects, mass transfer over deformable bubble interfaces, etc.) are still computationally too expansive to allow a reliable quantitative prediction of industrial scale equipment. However, CFD can provide prediction of crucial hydrodynamic parameters (axial dispersion coefficients, hold-ups, drop size distributions, etc.) and their trends in dependency of operation parameters and geometry.

By calibrating the simulation algorithm with experimental data from pilot scale experiments, CFD is used to optimize the internals design by a systematic variation of geometric ratios. For experimental validation the continuous phase flow fields of a Rotating Disc Contactor were recorded with Particle Image Velocimetry (PIV) and the residence time distribution was obtained by tracer experiments. Based on experimentally validated models and parameters continuous phase simulations combined with the Euler-Langrange framework were used to optimize the ratio of the column internals.

With the primary target of minimizing axial dispersion and maximizing intensity/size of the toroidal vortices, geometrical internals ratios were systematically altered in several simulation runs, finally leading to a novel simplified design without stator discs. Subsequent dual phase simulations (Euler-Euler model combined with population balances) were performed to evaluate stability of flow patterns in dual phase operation. Smaller droplets and narrower drop size distributions were observed for the optimized internals geometry, confirming intensification in terms of overall column performance.

Mass transfer experiments were performed to prove operation performance in terms of separation efficiency. Compared with the standard RDC design the height of one transfer unit (HTU) was found to be 40% lower in a pilot scale RDC100 with the simplified internals geometry.

Beside better operation performance the apparatus design without stator internals suggests applicability in complex separation tasks, such as downstream processing of fermentation broths, with explicit tendency to third phase formation in the compartments and the settling zone.