@phdthesis{,
    author = {Haase, Katharina},
    title = {Investigation of bubble dynamics in turbulent background flow},
    editor = {},
    booktitle = {},
    series = {},
    journal = {},
    address = {},
    publisher = {},
    edition = {},
    year = {2021},
    isbn = {},
    volume = {},
    number =  {},
    pages = {},
    url = {},
    doi = {},
    keywords = {multiphase flow, bubbles, piv, flow, optical measurements, turbulence},
    abstract = {This thesis presents an experimental study of free rising bubbles and bubbles in emulated turbulence to study the interaction between the flow structures and the movement of the bubble. Of particular interest is the bubble shape and shape oscillations, since these mechanisms are important for mass transport through the bubble surface. By understanding these interactions and using these results and measurement data, the interaction between the flow structures and the bubble can be qualified and the data can also serve as a basis for numerical studies of the phenomena, but especially for improving the efficiency of chemical reactors and other industrial plants. Here for example mixing plants or medical applications. The present dissertation is divided into three parts. First, different aspects of movement, such as the path and shape changes, are investigated. For this purpose, a new water channel was built, which allows nonintrusive optical measurements on a single bubble with and without background fluctuations. These background fluctuations are emulating swarms of bubbles and generate flow statistics similar to a swarm of bubbles. The created fluctuations are similar to those used in real industrial applications. This is achieved by different grids which are inserted into the water flow of the channel. One chapter of this thesis is dedicated to the qualification of the grids. The best results were obtained with a so-called free moving particle grid. These particles consist of spheres or ellipses of the desired bubble size, between 5 and 10 mm. These spheres are connected by a sewing thread only to the layer above. This allows the entire strand to move freely in the water flow. This emulates not only the shape of the swarm of bubbles, but also the movement of the bubbles. This makes it possible to induce a good model swarm of bubbles in the flow. Different strands (particle size, volume fraction and particle shape) were qualified with respect to flow statistics, energy spectra and probability density functions. It was found that a grid consisting of about 5mm ellipsoids with a density of 10% shows the best agreement with real swarms of bubbles. This was evaluated by comparison with real bubble swarms from literature data. With this model, individual bubbles in the swarm are suspended in the flow and further investigated with respect to shape oscillation and path changes. Such bubbles can be considered as moving in a swarm of bubbles. Before the bubbles in the emulated swarmcan be measured, the bubbles are examined in still water, not only to establish reference cases, but also to develop optical methods and evaluation algorithms to study shape and path oscillation. Individual bubbles were measured and the time-resolved shape, path and oscillation of the bubbles are recorded and reconstructed in 3D. The bubble size was varied between 2mm and 6mm to characterize all possible shape and path options. While it is known that in still water smaller bubbles show a zigzag path, larger ones follow a spiral path. Not yet reported is the bubble motion in a turbulent countercurrent flow and especially the interaction with the mentioned turbulent structures. In order to understand the shape oscillations, a new 3D reconstruction method is also introduced, which calculates the 3D shape of the bubble based on the shadow images alone. Compared to a 2D evaluation used in the literature, the 3D technique allows to study the bubble shape in an emulated background turbulence. While the 2D evaluation is sufficient for smaller, freely rising bubbles, it shows discrepancies for larger bubbles and especially for bubbles moving in an emulated turbulence. From the 3D reconstruction, the surface-volume ratio method is used to describe the shape oscillations. With the 3D evaluation it was made clear that the two frequencies used in the literature to describe the oscillations, f2_0 and f2_2, actually overlap in the 3D oscillation and can be measured as f_R and f_S. With these frequencies the surface-to-volume ratio is described. These frequencies show that the bubble in the emulated turbulence experiences a reduced oscillation frequency with increasing diameter, which could not be measured with the 2D approach. The emulated turbulence impedes the natural frequency and reduces it by a factor of 3. Furthermore, the path and movement of the bubbles and the influence of turbulence on them are investigated. The pressure gradients in the flow, caused by the velocity gradients, determine the path of the bubbles and push them into regions of low turbulence and high velocity. Furthermore, it could be shown that a critical turbulence level exists above which the bubbles are influenced by the turbulence and no longer follow their original path. This threshold was about 15% turbulence level. By adding tracer particles to the flow, the wake structures for freely rising bubbles and bubbles in the emulated turbulence were reconstructed. While the wake of a freely ascending bubble remains for more than 20 bubble diameters, the wake behind a bubble in the turbulence is accumulated in close proximity behind the bubble in a distance of about 3-4 bubble diameters. The wakes also becomes wider and the intensity is reduced as the imposed fluctuations quickly disperse the flow. Understanding the bubble dynamics in a swarmis especially important for mass and heat transport phenomena. To investigate this, a chemical tracer was used in collaboration with the LMU in Munich. This tracer allows to visualize the mass transport through the bubble surface and into the flow. With this tracer it could be shown that the transported species is mainly trapped in the core vortex areas of the wake. The reconstruction algorithm was also applied to this measurement and a three-dimensional representation of the wake could be reconstructed. With this reconstruction it would now also be possible to study the mass transport in the turbulent flow. However, this is an enormous experimental endeavor, which was no longer carried out in the context of this work.},
    note = {},
    
    school = {Universität der Bundeswehr München},
    
}