dc.description.abstract | Oscillatory Flow Mixing is a recent development in mixing technology which has evolved
over the past decade. It has a number of similarities to other mixing technologies,
particularly pulsed and reciprocating plate columns, but at the laboratory scale has
demonstrated a number of advantageous properties. These properties (such as control of
residence time distribution, improved heat transfer and predictable mixing times) have
been demonstrated at the laboratory scale for a wide range of different potential
applications, but until now there has been a lack of firm understanding and research into
how the technology could be scaled-up into an industrial scale process.
This thesis addresses the problem of scale-up in Oscillatory Flow Mixing. It reports on a
programme of experiments on geometrically scaled apparatus with the measurement of
residence time distributions and flow visualisation as the principal methods of
investigating the wide range of flow conditions that can be achieved by control of net
flow and of oscillatory conditions. Results from these investigations are interpreted as
axial dispersion coefficients and also compared with results obtained computationally
using a fluid mechanics approach to simulate flow fields and the injection of inert tracers
into those flow fields.
Significant clarification is reported concerning the analysis of axial dispersion
measurements using the diffusion model for which conflicting solutions were identified in
the literature. The development of a flow visualisation technique using fluorescent dye
streaklines is also reported. Using the latter technique stable manifolds in Oscillatory
Flow Mixing have for the first time been experimentally observed as well as a range of
other flow regimes.
The study of scale-up was extended by the successful construction and investigation of an
alternative reactor geometry with the potential for use in large-scale plant.
From the work presented in the thesis it is concluded that Oscillatory Flow Mixing is a
technology which in general lends itself readily to scaling-up from laboratory to pilot
plant scale, and most probably to industrial scale. Experiments performed on small
laboratory apparatus (containing less than one litre of fluid) can with confidence be used
to predict mixing behaviour in much larger plant (containing hundreds of litres of fluid.) | |