Local Hydrodynamics, Mixing and Mass Transfer in Bubble Columns with Internals


Local Hydrodynamics, Mixing and Mass Transfer in Bubble Columns with Internals

Möller, F.; Macisaac, A.; Seiler, T.; Hampel, U.; Schubert, M.

Abstract
Bubble column reactors (BCRs) are apparatuses of choice for the chemical process industry due to their excellent heat and mass transport as well as their simple manufacturing and design without any moving part. They are commonly used for reactions, such as Fischer-Tropsch and methanol synthesis. Most of the reactions carried out in these devices are of exothermic nature and require an efficient heat removal, which can be achieved via internally placed tube bundle heat exchangers. For the Fischer-Tropsch synthesis, a specific heat exchanging surface area between 30 and 40 m²/m³ is needed, which can be provided, for instance, with dense tube bundles inserted in the reactor. Accordingly, the hydrodynamic behavior of the reactor is strongly affected and subsequently, mixing and mass transfer processes are altered, too. Hitherto, the influence of the most common configurations used for the design of shell and tube heat exchangers, i.e. triangular and square pitches, has not been subject of a comparative analysis. Therefore, a study was conducted to reveal global and local hydrodynamics in the equipped bubble column covering a comprehensive sub-channel analysis as well as to disclose the pattern’s influence on gas-liquid mass transfer and liquid mixing behavior.

Figure 1: a) Bubble column reactor with internals, b) with spacer layouts, and c) summary of the studied tube bundle configurations.
A bubble column reactor with an internal diameter of 0.1 m (ID) and 2 m height (clear liquid height of 1.1 m) was employed and operated with deionized water and air. For the internal tube bundles, two configurations, namely, triangular and square tube pitch of two tube diameters (8 and 13 mm), ensuring a similar cross-sectional coverage were chosen (Figure 1).
The ultrafast X-ray tomography was used to disclose the hydrodynamic parameters bubble size and gas holdup. In particular, characteristic sub-channels along the reactor diameter were analyzed (Fig. 2a) to study the effect of the sub-channel position compared with the cross-sectional averaged data in order to examine whether a single sub-channel analysis is sufficient to describe the whole bubble column behavior. In addition, a fast-responding oxygen needle probe was used for mass transfer measurements (Fig. 2b) and wire-mesh sensors (WMS) were applied to track the dispersion of conductive tracers throughout the column’s cross-sectional area (Fig. 2c). The study covered homogenous and heterogeneous flow conditions (2 – 20 cm s-1) in order to also reveal the influence of different regimes.

Figure 2: a) Sub-channel flow structure, b) Mixing analysis based on 2D tracer distributions and c) mass transfer evaluation via oxygen saturation.
The hydrodynamic behavior within the sub-channels was found to depend strongly on their position within the reactor’s cross-sectional area. Furthermore, the influence of tube pattern is also very pronounced. The square configurations show advantageous hydrodynamic behavior with regard to the gas-liquid mass transfer. The small bubbles of narrow size distribution are evenly-distributed over the cross-sectional area and rise with lower velocity, thus, resulting in higher gas holdup and longer bubble-liquid contact time. Triangular configurations, however, introduce a strong flow resistance for the moving bubbles and cause flow asymmetries.
The mixing behavior in the columns with internals was quantified considering axial and radial dispersion of the tracer (Forret et al., 2003). The resulting axial dispersion coefficient was subsequently used for the estimation of the overall volumetric mass transfer coefficient. The dispersion coefficient depends strongly on the internals’ geometry as well as on the bottom structure, i.e. mixing is strongly increased when U-tube bottom ends are used. The mass transfer, however, is lower, with respect to an empty BCR, when internals are inserted due to turbulence dampening.
References
A. Forret, J.-M. Schweitzer, T. Gauthier, R. Krishna and D. Schweich, Liquid Dispersion in Large Diameter Bubble Columns, with and without Internals, The Canadian Journal of Chemical Engineering, Volume 81, June-August 2003, 360-366

  • Lecture (Conference)
    13th International Conference on Gas–Liquid and Gas–Liquid–Solid Reactor Engineering (GLS-13), 20.-23.08.2017, Brussels, Belgium

Permalink: https://www.hzdr.de/publications/Publ-25049
Publ.-Id: 25049