Search

This paper reports an experimental study of the mixing of a passive scalar in a bubble column at high Reynolds number and average gas volume fractions ranging from 2.0 % to 7.5 % . Starting from a homogeneous bubble column, the bubbly flow is progressively destabilized by imposing a gradient of gas volume fraction at the bottom of the tank. In that way, a single recirculation is produced, which allows to investigate the impact of a large-scale buoyancy-driven flow on the mixing of a passive scalar. It is shown that, as long as the shear-induced turbulence generated by the recirculation is negligible, mixing results from two main mixing mechanisms: the transport by the mean liquid velocity and the mixing induced by the bubbles. While the transport by the liquid recirculation can be accounted for by an advection term, the mixing induced by the bubbles is a diffusive process, the effective diffusivity of which has been measured in a homogeneous bubble column by Alméras et al. (2015). However, once the shear-induced turbulence produced by the shear develops, its role upon the mixing has to be taken into account too. [ABSTRACT FROM AUTHOR]

The present contribution reports an experimental study of the mixing of a passive scalar of very low diffusivity in a homogeneous swarm of inertial bubbles rising in a thin gap. A patch of fluorescent dye is injected within the swarm, and we observe the evolution of its mass in a given region of observation. We analyse the effect of the liquid agitation on the mixing mechanisms varying the gas volume fraction from 1.3 to 7.5 %, while the Reynolds number of the bubbles, Re = 450, their Weber number, We = 0.7, and the gap-to-bubble diameter ratio, w/ d = 0.25, are kept approximately constant. Here, the in-plane local mass of dye is measured by using a two-dyes planar laser-induced fluorescence (PLIF) technique that has been adapted to fix the problem of multiple light reflections at the bubble interfaces. Indeed, they induce both temporal and spatial variations of the captured light intensity that are superimposed to the effective fluorescence signal and prevent from using a standard PLIF technique. The analysis of the instantaneous concentration fields reveals the dominant role of the bubble wakes in the scalar transport. It is shown that mixing in this planar confined geometry is very efficient and enhanced by the increasing gas volume fraction. The present study also highlights that the mixing is not governed by a Fickian law of diffusion. [ABSTRACT FROM AUTHOR]

The present work investigates the mixing of a low-diffusivity dye in a swarm of bubbles at high Reynolds number confined in a Hele-Shaw cell for gas volume fractions ranging from 1.4 to 5.4%. A patch of a fluorescent dye is injected within the swarm and, during its mixing, its concentration is measured at a given location in an observation volume of 4.5 mm 2 by means of Laser Induced Fluorescence at a frequency of 250 Hz. A spectrometer is used to analyse the light issued from the observation volume and to distinguish the fluoresced light from other light sources. Simultaneously, the bubble distribution around the observation volume is imaged with a high speed camera synchronised with the spectrometer in order to assess the LIF technique in bubbly flow. Thanks to the good time resolution, rapid and intense concentration fluctuations corresponding to dye patches passing through the observation volume are recorded and are superimposed to a slow global evolution. This slow global evolution shows first an increase of the concentration and then an exponential decrease due to the mixing by bubble-induced agitation. This exponential decay, which is incompatible with a diffusion process, is consistent with the transport by dye capture in bubbles wakes that are quickly dampened by the shear-stress at the walls. The one-point statistics of the concentration fluctuations (probability density function and spectrum) also point out that mixing in a confined bubbly flow is intermittent and convective. [ABSTRACT FROM AUTHOR]