Investigation of microalgae and bubble interaction in electroflotation via image processing

Microalgae are becoming increasingly important for numerous applications such as food or pharmaceutical products. Flotation is an effective and comparatively inexpensive process for dewatering of the algal biomass after cultivation. In electroflotation, the hydrophobic algal cells attach to the surface of rising gas bubbles generated by water electrolysis and can be removed as a concentrated froth. For enhanced floatability, the size of microalgae can be increased by flocculation, e.g., with chitosan. Chitosan is a non-toxic, non-contaminating biopolymer that has proven to be a practical flocculant for microalgae. The effectiveness of the flotation process is influenced by numerous variables. At the same time, the mechanisms of the attachment of the algae to the bubbles are not fully understood. Hence, the aim of the presented work is to gain a deeper insight into the processes involved in the electroflotation of microalgae, like the algae-bubble-interaction, using optical measurement methods and machine learning (ML) based image processing.
A main focus is on the number and size of bubbles generated by electrolysis, as well as the size of Chlorella vulgaris agglomerates created by flocculation with chitosan. The properties of the bubbles were influenced by changing the electrolysis voltage and evaluated by image processing methods on microscopic images. Using laser diffraction spectroscopy, the influence of different chitosan dosages and flocculation times on the agglomerate size were analyzed. The size distribution is found to depend strongly on the varying biological properties of the microalgal suspension. Nevertheless, some general recommendations for an optimal chitosan concentration range could be deduced. In order to identify conditions promoting a successful attachment of algae to bubbles, an ML based method using series of microscopic images for visualization of the rising bubble and agglomerate paths during bubble-algal interaction was developed. The results show that a similar size of bubble and microalgal agglomerate is beneficial for enhanced bubble-algae interaction. For the analyzed voltage range, the mean bubble size was approximately 20 μm. The flocculation experiments showed that agglomerate sizes of 20 μm or higher are also achievable and thus, the microalgae flocs can be tuned to a well-floatable size range. Summing up, it was possible to derive first conclusions on how to promote effective electroflotation of microalgae. The developed visualization method contributes to a better understanding of flotation mechanisms and can be used as a basis for further research.