Skip to main content
eScholarship
Open Access Publications from the University of California

Real-Time Monitoring of Reverse Osmosis Membrane Integrity

  • Author(s): Surawanvijit, Sirikarn
  • Advisor(s): Cohen, Yoram
  • et al.
Abstract

Reverse osmosis (RO) membrane desalination is the primary technology for seawater and brackish water desalination, agricultural drainage desalting, as well as municipal wastewater recycling for potable water reuse applications. RO membranes achieve high salt rejection (>95%) and in principle should provide a complete physical barrier to nanosize pathogens (e.g., waterborne enteric viruses). However, in the presence of imperfections and/or membrane damage, membrane breaches as small as 20-30 nm in diameter can allow nanosize pathogens to pass through the membrane and contaminate the product water stream. In order to demonstrate that the level of removal of these contaminants by RO processes will assure rigorous public health protection, there is a need for real-time RO membrane integrity monitoring (MIM). However, at present, reliable real-time RO membrane MIM methods are unavailable for the detection of membrane or module breaches.

Given the above needs, a Pulsed Marker Membrane Integrity Monitoring (PM-MIM) approach was developed for real-time assessments of RO membrane integrity, based on characterization of fluorescent molecular marker passage across an RO membrane. This approach involves monitoring the dynamic change in marker concentration in the RO permeate (product) stream in response to a pulsed marker injection into the RO feed stream. The presence and characteristics of membrane integrity loss is identified by decoupling marker diffusive and convective transport through the RO membrane. The approach was first evaluated the passage of a molecular marker through intact and compromised RO membranes (with micron-size membrane breaches) in a bench-scale, plate-and-frame (PFRO) system. It was demonstrated that while marker passage through intact RO membranes was governed by diffusive transport, enhanced marker passage for compromised membranes was associated with convective transport through the breached areas. Therefore, detection of enhanced molecular marker passage was indicative of membrane integrity loss.

One of the potential causes of RO membrane integrity loss in RO plants is oxidation of the polyamide RO membrane surface by chlorinated disinfectants, which are typically introduced into the RO feed stream to prevent biofouling. The suitability of the PM-MIM approach for quantification of the extent of membrane integrity loss due to oxidation was evaluated for flat-sheet RO membranes that were degraded via exposure to sodium hypochlorite (NaOCl) solution at various exposure conditions (i.e., NaOCl concentration and exposure time). Upon membrane exposure to 50-200 mg/L NaOCl for a period of 2.5 to 10 hours, there was a significant increase in marker solution-diffusion and convection across the RO membrane as a result of changes in membrane surface physicochemical properties (surface roughness, chemical composition, and wettability) and structural damage (as evident from the presence of micron-size breaches). The severity of membrane integrity loss, represented by an equivalent breach size, increased with chlorine feed concentration and exposure time, with the latter having a more pronounced impact on membrane integrity.

In order to implement the PM-MIM approach in full-scale RO plants, which typically consist of multiple spiral-wound RO (SPRO) elements, an analytical framework was developed for characterization of marker passage through RO membranes in each individual SPRO element in multi-element SPRO systems. The approach is based on monitoring and analyzing the dynamic change in the marker concentration in the combined permeate stream (from multiple SPRO elements), in response to a pulsed marker injection into the RO feed. Subsequently, marker convective transport, across the RO membrane in each SPRO element, was be resolved via data fusion of dynamic marker response with online water flow, feed salinity, and pressure data, combined with the estimated marker residence time for each SPRO element. Experimental evaluation, which was carried out with intact and compromised SPRO elements (with mechanically induced membrane breaches), demonstrated enhanced marker passage for compromised SPRO elements, with increased marker passage that correlated with increased breach size. The PM-MIM approach was shown to be both technically and economically suitable for multi-element Ro system and through the proposed protocol for its implementation can serve to quantify the severity and identify the approximate location of membrane integrity loss in a multi-element SPRO plant.

Main Content
Current View