Analysis of mixing, mass transport and complex reactions in gas-liquid flows by multiphase DNS

Many important processes, both natural and man-designed, take place in a gas-liquid environment. In these systems, the gas exists in the forms of bubbles, which under the influence of a variety of forces (among which buoyancy is the most common) propagate through the liquid phase. Depending on various system parameters the ensuing multiphase flows can vary dramatically. On the macro-scale, the flow can vary from a homogenous bubbly flow to a violently agitated churn-turbulent one; from a flow exhibiting periodically oscillating bubble plumes and steady vortical structures to a slug flow, in which horizontal layering of the two phases occurs. On the micro and meso-scales, the properties of the individual bubbles, such as shape and size, can lead to the formation of qualitatively disparate wake flows.; One main consequence of the existence of this multitude of flow regimes is that the continuous phase mixing and mass transport will vary accordingly. It is well known that for many reactive systems, mixing can have a significant impact on the final product distribution. As a corollary, the selectivity of many reaction networks toward a given product will change depending on the hydrodynamic behavior of the multi-phase system. Since such variability can be highly undesirable in industry, an in-depth understanding of the hydrodynamics of such systems and their influence on reactive processes is necessary for the proper design and operation of multiphase reactors. Since the experimental study of gas-liquid systems is both expensive and strongly limited by technical issues, numerical simulations were performed for the purposes of our investigation. To avoid the inherent unreliability of empirical models, all simulations were based on first principles.; The approach adopted in this work was one of successive approximations. A simplified model, describing an ideal situation was used to study the basic phenomena in the system. This model was then gradually extended to represent progressively more realistic situations by relaxing many of the initial assumptions. Finally, as an application of tools developed, complex, real world systems were simulated, such as the gas-liquid reactive flows encountered in liquid-phase hydrogenations and in industrial bioreactors.