The flow of pulp suspensions
Large quantities of pulp suspensions are handled in the pulp and paper industry and knowledge of their flow behaviour is very important in the design and operation of process equipment. Despite this, there is still a lack of a fundamental understanding of the mechanisms of the pulp suspension flow. The suspension consists of fibres and water, and the fibre phase causes the complex flow behaviour due to the characteristic property of the fibres to aggregate and form coherent networks. Moreover, there seems to be areas with lower amounts of fibres near solid walls as a result of a compression of the network and orientation of the fibres, and this also affects the frictional behaviour of the pulp suspension.
In this thesis the flow of pulp suspensions in pipes has been studied experimentally using two different techniques, Particle Image Velocimetry (PIV) and Ultrasound Velocity Profiling (UVP). The experimental data from the PIV study and the LDA study by Pettersson et al. (2006) were further analysed to get a more detailed understanding of the mechanisms of the plug flow of pulp suspensions. Computational Fluid dynamics (CFD) was used to simulate the flow in an agitated pulp stock chest using a rheology model, where the local fibre concentration is taken into account.
Visual observations of images acquired in the PIV study were made to couple the structures in the flow to the flow regimes and related pressure drop curves reported in the literature. The images show that the distribution of fibres is very inhomogeneous in the near wall region, especially at low flow velocities, while the flow becomes much more structured when the velocity is increased. The mean velocity profiles show a distinct plug flow region at the centre of the pipe, and there is a steep velocity gradient close to the walls. The plug region increases with an increase in bulk concentration or a decrease in bulk velocity. This is consistent with previous studies. Both experimental techniques (PIV and UVP) can be used to measure the velocity profile, although the results of the two techniques show some deviations. Both techniques indicated the existence of a concentration profile near the wall, but the effects of various flow parameters are not clear. Further, the velocity and pressure drop data were used to estimate the shear stress at the wall, apparent boundary layer viscosity, equivalent fibre concentration at the wall, and area fraction of solid-wall contact. The results indicated two-phase flow effects near the wall, and the order of magnitude of the obtained values seems realistic.
The results of the CFD simulations show how the change in the local fibre concentration affects the cavern formation, thus also affecting the mixing efficiency of the tank. The flow patterns obtained in the transient simulations, including the fibre concentration variation, indicate that there are areas that will remain unchanged over time.
Particle image velocimetry
Computational fluid dynamics
Ultrasound velocity profiling