Microbial community analysis in biologically active filters exhibiting efficient removal of emerging contaminants and impact of operational conditions

Abstract

In biologically active filters (BAFs), microorganisms acclimated on the media surface are the key players responsible for removing organic water contaminants. In this study, next generation sequencing by Illumina MiSeq was used to characterize the microbial community structures in the influent, effluent, and media of a set of bench-scale BAFs that have been demonstrated with high removal efficiency (>75%) of 16 contaminants of emerging concern (CECs), which include a variety of pharmaceuticals (e.g., sulfamethoxazole and ibuprofen), X-ray contrast agent (i.e., iopromide), and pesticides (e.g., atrazine) that are prevalently found in municipal waste streams. Proteobacteria and Planctomycetes were the most abundant phyla in filter media, while the influent and effluent samples were dominated by Proteobacteria, Actinobacteria, and Chlamydiae. Factorial and principal component analysis revealed microbial structures in the media were significantly affected by the operation conditions, including media type (GAC versus dual media anthracite sand), EBCT (10 versus 18 min), and pre-ozonation. Detrended correspondence analysis demonstrated media materials predominantly governed the structures of the acclimated biofilm in BAFs as they provide direct attachment surface. This is in line with the higher microbial activity and better treatment performance exhibited by GAC BAFs compared to the dual media BAFs, corroborating the importance of filter media selection to promote the acclimation of active and robust biofilm for efficient CEC removal. Principal component analysis revealed the significant influence from ozonation, which does not only break down CECs, but also stimulates microbes that grow on the ozonation products. Partial canonical correlation analysis further proved the shaping of biofilm communities on the BAF media is more associated with media type and ozonation compared to EBCT. Putative CEC degraders are predicted based on their dominance in the media and degradation capabilities reported in previous literature. This is the first study to examine the relationship between the microbial community structure and the BAF operating parameters, which are both aligned with the treatment performance exhibited by the BAFs.

Keywords: Biological active filters; Contaminants of emerging concern; Empty bed contact time; Granular activated carbon and dual media; Microbial community structure; Pre-ozonation.