Fluid flow and mass transfer studies in mixing processes

Al-Najjar, Shahad Zuhair Atta (2019). Fluid flow and mass transfer studies in mixing processes. University of Birmingham. Ph.D.

Al-Najjar2019PhD_Redacted.pdf
Text - Redacted Version
Restricted to Repository staff only until 29 June 2030.
Available under License All rights reserved.
Download (9MB) | Request a copy

Abstract

Mass transfer coefficients in agitated solid-liquid systems are often estimated from a Frössling-type correlation which depends on the particle slip velocity. In a stirred suspension, the slip velocity will be a function of the hydrodynamic conditions prevailing in the vessel as well as particle size and will vary from point to point. Attempts to measure the slip velocity or estimate it based on Kolmogorov’s theory have not been successful. Therefore, for such calculations, it has been customary to take the free terminal settling velocity of a solid particle, as a representative of its mean slip velocity but the possible error involved has never been quantified, however. Using experimentally based estimates of local slip velocity obtained in a model solid-liquid system of mechanically suspended glass beads in water, we hereby assess the order of magnitude of the error likely to be incurred when using the particle settling velocity for mass transfer calculations.

The local 3D velocity components of both, the solid phase and the liquid phase were measured as a function of spatial position in a stirred vessel, using the technique of Positron Emission Particle Tracking (PEPT). The Lagrangian data were analysed using a MATLAB code to obtain a complete Eulerian description of the solid-liquid flow including time-averaged estimates of slip velocities of the solid particles.

Three different suspensions were studied under the just-suspended regime: mono-disperse of 3 mm diameter particles, binary-disperse of 1 and 3 mm particles, and poly-disperse consisting of five different sizes of particles ranging from 1 to 3 mm. Experiments were conducted at four different total solid mass concentrations: 5.2 wt%, 10.6 wt%, 20 wt% and 40 wt%. In the binary and poly-disperse suspensions, the different size fractions were mixed in equal proportions. Agitation was achieved by a 6-Blade 450 pitched-turbine operating in either up-pumping or down-pumping mode.

The azimuthally-averaged error between the local mass transfer coefficient based on the true slip velocity and that based on particle settling velocity were plotted as radial and axial profiles. Results showed that the local mass transfer rate varied considerably throughout the vessel and the use of the settling velocity can lead to extremely large errors in the estimation of the local mass transfer coefficient. The error of using right slip velocity can reach approximately three times in comparison with considering terminal settling velocity and be highest in the impeller region and near the wall. The largest particles were the most affected by the degree of poly-disperse of the suspension and exhibited their largest mass transfer errors in the suspension with five particle size fractions.

The mass transfer behaviour in mixing processes, specifically related to solid-liquid mixing systems concerning dissolved particles in water, has been studied in this work. A deep understanding of the mass transfer phenomenon is provided by the discussion of different correlations used to calculate the overall mass transfer coefficient and compare it with the existing methods. A novel approach is adopted with a unique experimentation method in positron emission particle tracking (PEPT) for solid-liquid mixing systems. This work attempts to track solid phase particles (i.e. dissolved particles by making one of them a radioactive with a tracer injected to its core). Hence, these experiments can characterize the mixing process that takes place inside the tank and to the visualize the overall mixing process, which is concerned with the behaviour of solid particles during the dissolution process of the solid mass transfer phenomena. In terms of the particles, these were prepared specially to match the size and density of plastic particles (modal particles) in order to compare the flow patterns between the two; i.e. using dissolved particles once and for the second time, using the plastic particles under the same experiment characterizations to assess the overall quality of mixing processes.

Two-phase (solid-liquid) and three-phase (solid-liquid-gas) stirred tank reactors to constitute a major class of multiphase reactors. Such types of reactors are extensively employed in the chemical industries due to their steadfastness of operation and flexibility. Agitation is mainly applied to make sure that all available surface area is used for the process of mass transfer. A study of the mass transfer behaviour in an agitated solid-liquid system was performed to compare with the results of the mass transfer behaviour in an agitated solid-liquid system that are studied at constant experiment conditions in terms of the concentrations of solid and the regime of mixing. while the only difference is the existence of gas flow rate, which was provided to the system by a pipe sparger. It was revealed from the findings of the comparison that to achieve same conditions for performing experiments, the ‘just suspended’ speed needs to be higher in sparged gas experiments than the other one without gas flow rate.

The internal flow field and spatial dispersion of floating particles have been scarcely studied experimentally. Owing to the opaque nature of such suspensions, work has been restricted to flow visualizations only through the wall and measurement of cloud depth. The liquid and solid phase velocity distributions allowed estimations of the spatial distribution of particle-fluid slip velocities in the vessel which were used to assess the errors involved in local mass transfer predictions using Frössling-type correlations.

Type of Work:

Thesis (Doctorates > Ph.D.)

Award Type:

Doctorates > Ph.D.

Supervisor(s):