Use this URL to cite or link to this record in EThOS:
Title: Gas-liquid mixing in stirred vessels imaged using electrical resistance tomography (ERT)
Author: Forrest, Andrew E.
Awarding Body: University of Manchester : UMIST
Current Institution: University of Manchester
Date of Award: 1997
Availability of Full Text:
Access through EThOS:
A considerable effort has been invested over the past forty or so years in attempting to thoroughly understand the basic mechanisms which govern the dispersion of gas inside mechanically stirred mixing vessels. Previous research has concentrated on visual flow studies of gas-liquid flow structures and on measurement of overall properties such as agitator power consumption and gas hold-up. Any local measurements have usually been obtained intrusively and are point-wise giving limited information about the internal flow regime. The evolution of tomographic techniques for non-medical applications now offers a powerful method of obtaining a measurement based knowledge of the mixing processes which occur inside stirred tank reactors. This thesis addresses the use of Electrical Resistance Tomography(ERT) to quantify the mixing behaviour of a gas-liquid (air-water) system inside a plant scale stirred tank reactor. The technology from previous ERT applications was used to develop an 8-plane, 16- electrode measurement system which was installed inside a 2.3 m3 polypropylene tank. The existing Mk lb data acquisition system (DAS) was extended to allow almost instantaneous collection of data from all 8 electrode planes. The tank had a standard geometry with four full length baffles and a six-blade Rushton turbine. Qualitative and quantitative images of the internal, cross-sectional conductivity distribution were reconstructed from the ERT system measurements. 3-D solid body images which illustrate the effects of increasing stirrer speed on the internal gas-liquid. flow regime were constructed from the conductivity distributions obtained in all 8 planes. An existing network-of-zones (NWZ) model of gas-liquid mixing in stirred tank reactors was modified to allow direct comparison with the ERT images. The model was also used to quantify the inherent errors present in the quantitative, modified Newton-Raphson (MNR), reconstruction algorithm. From work carried out in conjunction with BNFL Fluorochemicals further reactor models are presented for the absorption and reaction of fluorine gas which incorporate the concept of interconnecting back-mixed elements/zones without making prior assumptions on the role of film wise and bulk wise reaction. In general the ERT images and the corresponding NWZ model's predicted gas distributions compare well with gas-liquid flow patterns obtained from visual flow studies. In particular, the solid body images describe the flow structure well and confirm that the spatial gas dispersion for a Rushton turbine is far from homogeneous. The similarities between the ERT images and the NWZ model are discussed at two different operating conditions. This illustrates the utility of ERT to verify and develop computational-based fluid dynamics( CFD) predictions. The sensitivity and spatial resolution of the ERT system were assessed using a specially constructed gas-liquid phantom. A full analysis was made of some experiments on the absorption of fluorine from a gas stream chain bubbled through a column of fomric/sulphuric acid. This analysis was used as a basis in developing two further reactor models for simplified co- and counter-current flow using interconnecting back-mixed elements. These models can predict not only the depletion of the reagents from the appropriate phases, but also how the relative amounts of film and bulk reaction change through the reactor. This approach needs further development for application to stirred vessels. It is an important element in properly understanding the chemical selectivity of a stirred gas-liquid reactor. The work demonstrates the ability of ERT to fully describe gas-liquid mixing processes inside a plant scale vessel. The potential use of ERT in the design, monitoring and control of industrial mixing vessels has been realistically assessed. The major areas where ERT needs improvement for ultimate use in industry are given in the recommendations. The necessity to extend the absorption and reaction models developed in this work to include complex reaction schemes is also discussed.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available