Combining reverse osmosis and microbial degradation for remediation of drinking water contaminated with recalcitrant pesticide residue. Mem2bio

Groundwater contamination by recalcitrant micropollutants such as pesticide residues poses a great threat to the quality of drinking water. One way to remediate drinking water containing micropollutants is to bioaugment with specific pollutant degrading bacteria. Previous attempts to augment sand filters with the 2,6-dichlorobenzamide (BAM) degrading bacterium Aminobacter niigataensis MSH1 to remediate BAM-polluted drinking water initially worked well, but the efficiency rapidly decreased due to loss of degrader bacteria. Here, we use pilot-scale augmented sand filters to treat retentate of reverse osmosis treatment, thus increasing nutrient availability and residence time in the biofilters. In a first pilot-scale experiment, BAM and most of the measured nutrients were concentrated 5-10 times in the retentate. This did not adversely affect the inoculated bacteria and the general prokaryotic community of the sand filter presented only minor differences. On the other hand, the high degradation activity was not prolonged compared to the filter receiving non-concentrated water at the same residence time. Using laboratory columns, it was shown that efficient BAM degradation could be achieved for >100 days by increasing the residence time in the sand filter. A slower flow may have practical implications for the treatment of large volumes of waters, however this can be circumvented when treating only the retentate water equalling approximately 10% of the volume of inlet water. We therefore conducted a second pilot-scale experiment with two inoculated sand filters receiving membrane retentate operated with different residence times (22 versus 133 minutes) for 65 days. While the number of MSH1 in the biofilters was not affected, the effect on degradation was huge. In the filter with short residence time, BAM degradation decreased from almost 90% to a stable level of 10-30% degradation within the first two weeks. The filter with the long residence time initially showed >97% BAM degradation, which only slightly decreased with time. Besides being the first successful pilot-scale study with bioaugmented sand filters for pesticide residue degradation, our study demonstrates the advantage of combining membrane filtration with bioaugmented filters in cases where flow rate is of high importance.