PhD defence: From Pores to Pilot Filters: Biofilm-Driven Bacterial Removal in Slow Sand Filtration

Clean and safe drinking water depends on treatment systems that can reliably remove microorganisms. One of the oldest and most robust methods for this is slow sand filtration. In these filters, water slowly passes through a sand bed. Over time, a living biological layer develops on top of the sand, known as the Schmutzdecke. This layer contains microorganisms and sticky biological material, and it plays an important role in removing bacteria from water. However, exactly how this biofilm improves removal, and how its development can be predicted, has not been fully understood.

This thesis investigated how biofilms control bacterial removal in slow sand filters, from the scale of individual pores to pilot-scale filters. Transparent microfluidic models were used to directly observe how particles move through pores and interact with biofilms. These experiments showed that biofilms change the pore structure by creating rough surfaces, narrow passages, dead-end zones, and preferential flow paths. These changes strongly increase particle and bacterial retention.

Laboratory and pilot-scale slow sand filters were then used to study how these mechanisms appear in real filtration systems. The results showed that the top layer of the filter becomes the main removal zone as the filter matures. The removal performance was not explained well by filter age or biomass alone. Instead, the biochemical composition of the Schmutzdecke, especially the protein and carbohydrate content of extracellular polymeric substances, was a better indicator of bacterial removal.

Overall, this thesis shows that bacterial removal in slow sand filters is controlled by biofilm structure, hydraulic changes, and biofilm chemistry. These insights can help improve predictive models, monitor filter maturity, shorten ripening periods, and optimise drinking water treatment.