Bubble columns are used in the chemical, petrochemical, biochemical and metal industries. Power-to-gas approaches are also gaining importance due to the decentralization of electricity grids. For example, methanation produces synthetic methane (CH4) from hydrogen (H2) obtained by electrolysis and carbon dioxide (CO2) or carbon monoxide (CO). A three-phase system in a bubble column is used. The two reactants are in gaseous form and the catalyst is in solid form. In addition, a liquid heat transfer medium is used. However, certain quality requirements must be ensured for the gas feed. Particularly in the case of reactive bubble columns, their efficiency is partly determined by the prevailing local hydrodynamics. Therefore, to increase yield and selectivity, a better understanding of the local hydrodynamics in bubble columns is required, which the state-of-the-art design based on integral methods cannot provide.

Despite the promising description of local hydrodynamics by CFD, some open questions remain that need to be solved for the predictive description of reactive bubble columns:

1. Influence of bubble interaction (e.g., bouncing).
2. Influence of bubble size distribution on hydrodynamics (mono-, polydispers).
3. Mutual interaction of reaction and hydrodynamics (bubble-induced turbulence).

Mass transport at the phase interface and in the bubble backlash is influenced by bubble interaction (hydrodynamic stress) in addition to local hydrodynamics. This includes bouncing and the resulting bubble dilation as well as hydrodynamic stress during coalescence and film drainage.
The measuring cell available at the institute allowed investigations on a hanging bubble as well as on freely rising bubbles by applying small amounts of liquid (<500 ml) and gas according to the generated number of bubbles under inert conditions. By automating the test cell, a high reproducibility of the measured values was achieved. Thus, using a Laser Induced Fluorescence system, it was shown that mass transfer increases briefly under hydrodynamic stress conditions until the interface regenerates.

The reactive mass transport was further investigated on a Fe(II)-ligand-NO system. It was possible to describe this well on the basis of two-dimensional simulations and to use the findings for a scale-up to a pseudo 2D bubble column.
For this purpose, an extension of an Euler-Euler approach to the description of polydisperse systems as well as to the description of the reactive mass transport and temperature development in the apparatus was carried out. It was shown that if the bubble size is known, the interaction between hydrodynamics and reaction can be well described by numerical simulation.
In the future this will allow to design efficient, sustainable and economic processes and to describe the mass and heat transfer in the best possible way by a combined experimental as well as numerical approach.