Safeguarding reverse osmosis membrane integrity in drinking water production: A pilot study using 2 native bacterial communities and viruses in surface water as integrity indicators

Abstract

Water treatment train including a reverse osmosis barrier is capable of producing high-quality water coupled with microbial safety. However, analytically proving this with log removal value (LRV) > 3 has remained a challenge. Conservative techniques that quantify salt or sulfate ion passage as virus surrogates are currently used to credit the membrane system for its microbial removal capacity. These methods are insensitive, and it is now understood that the transport mechanism correlation between these surrogates with the actual passage of viruses is weak. There is also a huge gap between the current log credit assigned (using spiking of surrogates in feed) and the actual log removal that can be proven in the drinking water plant. In this paper, we introduce two techniques that exploit the high concentration of microbial communities in fresh surface water by quantifying bacteria through flow cytometry (FC) and native viruses through qPCR assays. These two methods are compared to LRV of conductivity, TOC, and fluorescent markers. That high concentration of bacterial community allowed for LRV > 4 by flow cytometry in an intact module. After inflicting damage with a 1.5 mm defect, the FC method detected the damage, and a bacterial cell count >103 cell count/ml was measured in the permeate, while conductivity lacked sensitivity to monitor any integrity loss. Furthermore, after chlorine exposure to the membrane, both bacteria and virus detected no leakage, proving the membrane's intactness and also demonstrating the effect of chlorination on the passage of small solutes. The implementation of the two microbial quantifying methods in full-scale water treatment plants facilitates higher sensitivity in membrane integrity loss than currently used methods.