Modelling and simulation of two phase flows in bubble columns
Aeration of reactors is a wide used concept in all engineering discipline were heat and mass transfer, and thus mixing, play an important role. The term bubble column is used in general for such reactors. Through dispersing the gas in the liquid phase a very large surface is produced. As a result chemical reactions between the liquid and gas phase are intensified or even enabled. Through the buoyancy of the bubbles and the momentum exchange between gas and liquid strong unsteady liquid flows on different length scales occur. These turbulent flows are essential for mixing of the liquid phase inside the bubble column. The description of such systems is of interest of many studies in the last decades. For simulation of industrial scale problems the EulerEuler method is in general preferred. The drawback of this method is that the interaction between the phases has to be modeled completely. At this point the goal is to define a general closure model which covers all effects to ensure the predictability of the simulations.
The current focus of the work is on modeling and validating the momentum transport between the two phases. The models have to take into account complex phenomena like the interaction between the bubbles or the lateral bubble movement in a shear flow. Among others the mobility of the bubble surface plays an important role in in modeling these phenomena. The mobility is affected by impurities or surfactants which are connected with the appearance of the Marangoni effect. Such impairment of the surface mobility influence not only the bubble shape, they also lead to complete other behaviors of the bubble in shear flow or in the ascending process. To identify these effects is crucial for the next step, simulating chemical reactions where such surfactants could arise in complex substance systems.
Beside the small length scale the large length scales in the dimension of the reactor itself are very important for a predictive simulation of bubble columns. Through the pointwise gas injection complex flow structures occur which depend very strong from the reactor geometry. For example in Figure 1 can be seen a snapshot of a transient simulation of an uneven aerated bubble column. At this point one goal is to find a compromise between expensive transient simulation and large reactor geometries. Unsteady Reynolds Averaged Navier Stokes Equations (URANS) simulations are promising for simulating complete facilities while considering large scale fluctuations. In Figure 2 is the contribution of the resolved fluctuations to a normal Reynolds stress tensor compoenent shown, in Figure 3 the unresolved and total tensor component; the experiment is a table top bubble column conducted by bin Mohd Akbar , M. H., Hayashi , K., Hosokawa , S. & Tomiyama Multiphase Science and Technology, 24(3), pp. 197222.


Figure 1: Snapshot of a transient EulerEuler two phase simulation 
Investigations on the flow stability in bubble columns
Bubble columns are widely used for heterogeneous chemical processes. Their effectiveness strongly depends on the flow structure. Especially the transition between homogeneous and heterogeneous regime has been the focus of investigations for many years. The models and criteria, which can be found in the literature, are not able to explain the process in a consistent way. Nowadays, the bubble size dependent lateral lift force was identified as a key to solve this problem. For small bubbles this force has a stabilizing effect. For large bubbles, for which the lift force changes its sign it has a destabilizing effect. By means of a linear stability analysis this effect was investigated and criteria for the local stability of a bubbly flow dependent on the bubble size distribution were derived. The findings have been also checked by CFD simulations using the CFX10 software. Sources of local flow instabilities, which are characterised by a significant increase of the lateral velocity component, have been found in the regions predicted by the stability analysis. The flow behaviour is affected by interplay of local and global flow instabilities. After publication of the criteria for stability obtained by the linear stability analysis, these findings were confirmed experimentally by two groups (University of Montreal and University of Delft).
Figure 2: Scheme on the influence of the lift force on a 
Figure 3: Fields of the liquid velocity calculated with the CFDCode CFX from ANSYS in comparison with the obtained criteria for stability (from Lucas et al. 2006) 
Fundamental experiments
To support the theoretic work, experiments are realized in bubble columns. The focus of these experiments is to extend the closure relations for polydispersed bubbly flows in turbulent flow conditions. Also it is a goal to confirm the developed stability analysis with experimental data. Besides the experiments in highly purified water, surfactants and impurities are added to study the influence on bubbly flows. The second focus is on evolving measuring methods for mutliphase flows using using high speed cameras. Here, the pattern recoginition of bubbles (Figure 4), tracking of microbubbles (Figure 5), Particle Image Velocimetry (PIV) methods (Figure 6) and pattern tracking (Figure 7) are subjects of the current work.




References
Ziegenhein, T. 2016
Fluid dynamics of bubbly flows. Dissertation, TUBerlin.
Ziegenhein, T. & Lucas, D. 2017
Observations on bubble shapes in bubble columns under different flow conditions. Experimental Thermal and Fluid Science, (accepted).
Ziegenhein, T., Zalucky J., Rzehak, R. & Lucas, D., 2016.
On the hydrodynamics of airlift reactors, Par I: Experiments. Chemical Engineering Science, 150 , pp. 5465.
Ziegenhein, T.; Garcon, M. & Lucas, D., 2016.
Particle tracking using micro bubbles in bubbly ﬂows. Chemical Engineering Science, 153, pp. 155164.
Ziegenhein, T.; Rzehak, R.; Ma, T. & Lucas, D., 2016.
Towards a unified approach for modelling uniform and nonuniform bubbly flows, Canadian Journal of Chemical Engineering, 96(1), pp. 170179.
Ziegenhein, T. & Lucas, D., 2016.
On sampling bias in multiphase flows: Particle image velocimetry in bubbly flows. Flow Measurement and Instrumentation, 48, pp. 36–41.
T. Ziegenhein, R. Rzehak, D. Lucas
Transient simulation for large scale flow in bubble columns, Chemical Engineering Science, 2015, 122, 113.
T. Ziegenhein, R. Rzehak, E. Krepper, D. Lucas
Numerical Simulation of Polydispersed Flow in Bubble Columns with the Inhomogeneous MultiSizeGroup Model, Chemie Ingenieur Technik, 2013, 85, No. 7, 1080–1091.
T. Ziegenhein, D. Lucas, R. Rzehak, E. Krepper
Closure relations for CFD simulation of bubble columns, 8th International Conference on Multiphase Flow , ICMF 2013, Jeju, Korea, May 26  31, 2013
D. Lucas, E. Krepper, H.M. Prasser, A. Manera
Stability effect of the lateral lift force in bubbly flows, 6th International Conference on Multiphase Flow, ICMF 2007, Leipzig, Germany, July 9 – 13, 2007, paper S1_Mon_C_9
D. Lucas, E. Krepper, H.M. Prasser, A. Manera
Investigations on the stability of the flow characteristics in a bubble column, Chemical Engineering & Technology 29 (2006) 1066  1072
D. Lucas, H.M. Prasser, A. Manera
Linear stability analysis for the effect of the lift force in a bubble column, 7th German/Japanese Symposium on Bubble Columns, GVC, 20.23.05.2006, Goslar, Deutschland
D. Lucas, H.M. Prasser, A. Manera
Influence of the lift force on the stability of a bubble column, Chemical Engineering Science 60(2005)36093619
D. Lucas, H.M. Prasser, A. Manera
Investigations on the stability of a bubble column, Annual Report 2004, Institute of Safety Research, HZDR420, Rossendorf, March 2005, S. 16