Nano-plasmonic sensing for predicting fouling on a reverse osmosis membrane

The reuse of municipal wastewater is crucial to the development of new water resources, especially for agriculture. A challenge to the long-term sustainability of this approach is the presence of organic foulants in the feed water. While purification using a reverse osmosis (RO) membrane can effectively desalinate wastewater effluent to produce potable water, the main drawback is fouling of the membrane by the accumulation of a layer of organic matter from the effluent. Therefore, monitoring the propensity of pre-treated feed water to foul the RO membrane is essential for robust continuous RO operation. Electrical impedance spectroscopy (EIS), silt density index (SDI), turbidity measurement, and side stream membrane modules have been employed to predict fouling and enabling scheduled membrane cleaning. While superior RO fouling prediction capabilities were shown for EIS, other methodologies commonly provide quick but inaccurate assessments or accurate assessments at timescales too long to be useful in preventing fouling. This study investigated an innovative RO fouling prediction methodology, localized surface plasmon resonance (LSPR) sensing. We compared LSPR with predictions using SDI and a recently suggested quartz crystal microbalance with dissipation technique. The LSPR method showed high-sensitivity detection to model and environmental fouling agents by quantifying real-time foulant adsorption to the sensor surface. Our findings demonstrate that LSPR can surpass the traditional SDI method in predicting fouling propensity, likely owing to its high sensitivity to adsorbed material up to tens of nanometers from the sensor surface. LSPR thus offers a precise method of predicting RO membrane fouling that can potentially enable proactive fouling management, enhancing the longevity of membranes and reducing downtime during their operation. Synopsis Continuous wastewater reverse osmosis desalination ensures sustainable water resources, with fouling prediction via LSPR sensing vital for minimizing downtime and optimizing system efficiency.