Can we shape microbial communities to enhance biological activated carbon filter performance?

A new focus on biofiltration has emerged that aims to shape microbial communities to improve treatment efficacy. It is therefore necessary to understand the linkages between microbial community structure and biofilter function. However, the assembly and interaction of microbial communities in biological activated carbon (BAC) filters are unknown. In this study, we selected one coal-based granular activated carbon (GAC), GAC-13, with simultaneously developed micropore and micro-level macropore volume used for a bench-scale BAC column experiment, and compared it with other coal-based GACs and wood-based GAC in terms of the dissolved organic carbon (DOC) removal and microbial community characteristics. The results showed that there was no difference between the DOC removal efficiency of BAC-13 and the other two coal-based BAC filters with high iodine value in the period dominated by adsorption, while the DOC removal efficiency of BAC-13 (64.7±0.6%) was significantly higher than that of other BAC filters (36.3±0.8-54.1±0.4%) with a difference of 0.3-0.7 mg/L in DOC during the steady state. The bacterial communities were strongly assembled by deterministic rather than stochastic factors, where the surface polarity of GAC had a greater effect on the microbial communities than its physical properties. The corresponding co-occurrence network revealed that microbes in the BAC filter may be more cooperative than competitive. The keystone bacterium Hyphomicrobium, which had a relatively low abundance, contributed 0.3-1% more to the most abundant functions and produced 5-21 proteins/(g·GAC) more than the dominant bacterium Sphingobium. The metaproteomic-based approach could provide more accurate information regarding the contributions of different species to metabolic functions. The pore size distribution of GAC was found to be an important factor in determining BAC filter performance; the most important pore sizes were micropores and micro-level macropores (0.2-10 μm and >100 μm in diameter), and the latter impacted the abundance of keystone species. Overall, our findings provide new insights into shaping microbial communities by optimizing pore size structure to improve BAC performance, especially the abundance of keystone species.