Evaluation of Inline Characterization of Bubble Size Distributions in Highly Concentrated Dispersions

Gas-liquid mass transfer in bubble columns [1] and stirred vessels is a common mixing operation in the chemical, biochemical and mineral processing industries. Most often a component in the gas phase is transferred into the liquid where it reacts with a component to produce the desired product. In many cases the overall process will be mass transfer limited and the rate at which the reaction proceeds will be determined by the transfer rate of gas molecules into the liquid phase. There are also cases where a volatile component, for example a reaction by-product, must be stripped from the liquid phase [2].

Today many different techniques for sizing transient fluid particle behavior in production vessels are available. Often samples are withdrawn over time [3], which are later diluted or stabilized, prior to their measurements. These sampling techniques neither guarantee that the drop or bubble sizes are frozen, nor that they are preserved during the sampling and have been criticized by many authors. Especially for technical applications with fast reactions and strong coalescence such kind of measurements are not suitable. Even for sampling times less than one second, the drastic change in the flow condition during sampling results in significant measurement errors.

A state-of-the-art photo-optical inline particle measurement, combined with a neural network-based particle detection system, makes real-time bubble characterization during chemical processes possible (see fig. 1 for example results). In this project, a new prototype of the assessment of evaluation methods for image-based bubble measurement in chemical engineering is developed based on selected assessment parameters.

Moreover, different algorithms are tested for particle detection on several existing images from industrial and academic application examples. The results show the capabilities of inline imaging methods and compare accuracy and computation times of various bubble analysis algorithms. Based in this study, recommendations are provided as a guideline for choosing a suitable image analysis algorithm on specific particle detection problems. The imaging-based solution is used to study process engineering challenges; specifically the comparison of two different stirrer types for varying process conditions like stirrer speed and aeration rate will be discussed in detail.

Automated inline imaging techniques offer significant possibilities in process engineering in general and specifically for process control, design and scale-up tasks.

Figure 1 – graphical display of the bubble detection results in a typical double image (image with detection above, original image below) [4, 5]

Literature

(1) G. Besagni, F. Inzoli, and T. Ziegenhein, Two-Phase Bubble Columns: A Comprehensive Review. ChemEngineering, 2(2) (2018): p. 13.

(2) A. Hofmann, G. Schembecker, and J. Merz, Role of bubble size for the performance of continuous foam fractionation in stripping mode. Colloids and Surfaces A: Physicochemical and Engineering Aspects, 473 (2015): p. 85-94.

(3) C. Martínez-Bazán, J.L. Montanés, and J.C. Lasheras, On the breakup of an air bubble injected into a fully developed turbulent flow. Part 1. Breakup frequency. Journal of Fluid Mechanics, 401 (1999): p. 157-182.

(4) R. Panckow, S. Maaß, J. Emmerich, and M. Kraume, Automated Quantification of Bubble Size Distributions in an Agitated Air/Water System. Chemie Ingenieur Technik, 85(7) (2013): p. 1036-1045.

(5) R.P. Panckow, G. Comandè, S. Maaß, and M. Kraume, Determination of Particle Size Distributions in Multiphase Systems Containing Nonspherical Fluid Particles. Chemical Engineering & Technology, 38(11) (2015): p. 2011-2016.