RO membrane modules are very effective in removing viruses and salts, but concerns remain regarding their long-term integrity in full-scale RO spiral wound systems. Membrane defects can arise from delamination, chlorine, or abrasive particle exposure, and module defects can arise from permeate tube, O-ring, and glue line damage. These issues compromise the membrane and module's performance, allowing viruses to pass that are not detected by conductivity monitoring. This paper assesses the impact of damage on virus removal using biological indicators (natural virus markers (NV)) and non-biological surrogates (salt, sulfate, rhodamine WT, pyranine). Three categories of module damage were studied and characterized using different autopsy techniques: membrane module component damage (permeate tube damage, O-ring damage, and membrane delamination), oxidative membrane damage (by hypochlorite exposure dose), and membrane surface damage (by abrasive particle exposure). Module component damage resulted in increased water permeability and decreased natural virus rejection. The conductivity remained similar to that of an intact module except in the case of permeate tube damage. Chlorine exposure (9000 ppm.h) didn't result in detectable NV markers in the permeate, indicating an intact support layer despite active layer damage. Abrasive particle exposure by suspended silicon carbide in the feed resulted in scratches, observed by SEM images, and a decline in the rejection of fluorescent marker, which indicates membrane surface damage. Results demonstrate that NV markers most consistently determine virus removal capacity in modules compromised by component damage. However, for assessing active layer damage, solutes like rhodamine-WT and pyranine prove more effective. This underscores the necessity of employing multiple indicators for a comprehensive evaluation of membrane integrity.
Read full abstract