Modelling of mass transfer in Taylor flow: Investigation with the PLIF-I technique

•PLIF-I investigation is applied to Taylor flow in circular channel for the first time.•New data on mass transfer between bubbles and slugs in Taylor flow is obtained.•Model for separate contributions of bubble caps and lubrication films is proposed.•Model if separate contributions requires knowledg...

Full description

Saved in:
Bibliographic Details
Published inChemical engineering research & design Vol. 115; no. Part B; pp. 292 - 302
Main Authors Butler, C., Cid, E., Billet, A.-M.
Format Journal Article
LanguageEnglish
Published Rugby Elsevier B.V 01.11.2016
Elsevier Science Ltd
Elsevier
Subjects
Online AccessGet full text

Cover

Loading…
More Information
Summary:•PLIF-I investigation is applied to Taylor flow in circular channel for the first time.•New data on mass transfer between bubbles and slugs in Taylor flow is obtained.•Model for separate contributions of bubble caps and lubrication films is proposed.•Model if separate contributions requires knowledge of basic flow characteristics. An experimental investigation of mass transfer between bubbles, slugs and lubrication films is performed by means of optical techniques for air–water Taylor flows in a 3mm glass channel. Bubble size, shape and velocity, as well as slug size and film thickness, are measured by use of the shadowgraphy technique. The PLIF-I technique, with associated specific technical adjustments and image processing methods, gives access to the dissolved oxygen concentration in the liquid phase with high spatial and temporal resolutions. Values of concentration are measured in the slugs and, for the first time, in the lubrication films, leading to the quantification of mass transfer contributions from caps and from the central bubble body. Results for overall mass transfer are compared to literature models based on volumetric coefficient kLa. Among them the model proposed by Van Baten and Krishna (2004), which considers cap and film contributions to kLa. They are found to fail to accurately predict the experimental data of average O2 concentration in the liquid phase for short slugs (Ls<2dc). An improved version of this model is proposed, allowing calculation of O2 concentration for films and slugs along the channel. The model needs flow characteristics, which are obtained by shadowgraphy.
ISSN:0263-8762
1744-3563
DOI:10.1016/j.cherd.2016.09.001