Give Membranes The Virus Removal Credit They Deserve: Using Rapid In-Field Molecular-Based Methods (WRF 5209)

This project aims to develop cost-effective, rapid, in-field technologies for the quantification of viruses in reclaimed water. The first phase of the project focused on advancing the rapid in-field quantitative polymerase chain reaction (qPCR) technology for measuring viruses. The main objectives included reducing analysis time, evaluating impact of PCR inhibitors, determining recovery efficiency of each individual sample, and identifying suitable viral surrogates. Once these objectives were met, the enhanced method was tested at the Orange County (CA) Water District Groundwater Replenishment System (OCWD GWRS). During the period of sampling, the GWRS facility produced, on average, 115 MGD of high-quality water, with 79 MGD from microfiltration (MF) and 36 MGD from (parallel) ultrafiltration (UF) units. To compare the virus removal efficiency of MF versus UF units, samples of membrane feed (100-200 mL) and permeate (30-40 L) were collected from both MF and UF units. The samples were analyzed for norovirus GII (an enteric virus) and three viral surrogates (CrAssphage, PMMoV, and ToBRFV). To account for the virus losses during the analysis (concentration, extraction, and quantification), each sample was spiked with murine norovirus (MNV) as a matrix recovery control. The recovery efficiency of MNV in feed samples was determined to be 19.5% +/- 3.0% (n = 11), while in the permeate sample, it was 2.8% +/- 0.6% (n = 12). These recovery rates align with the acceptable range for virus recovery in water and wastewater samples. It is important to note that failing to adjust the final concentration of viruses based on these recovery efficiency values would lead to an overestimation of the LRV for viruses in the membrane filtration process (in this study, the potential overestimation is estimated to be 1 LRV). Among all the measured targets, ToBRFV exhibited the highest concentration in the membrane feed samples, with a value of 9.8 * 108 copies/L, followed by CrAssphage (1.1 * 108), PMMoV (1.0 * 108), and norovirus GII (3.4 * 105). In the MF unit, CrAssphage demonstrated the highest LRV of 2.8, followed by norovirus GII at 1.9, ToBRFV at 1.8, and PMMoV at 1.5. The UF unit showed superior virus removal efficiency compared to the MF unit, achieving the highest LRV of 4.2 for CrAssphage, followed by 3.9 for PMMoV, 3.7 for ToBRFV, and 3.1 for Norovirus GII. According to the obtained results, the UF unit achieved a minimum LRV advantage of 1.2 relative to the MF unit. When comparing the LRV of different targets in the MF and UF units, it is evident that LRVs do not exhibit consistent patterns. This discrepancy suggests that virus removal in membrane filtration is influenced by various factors such as membrane characteristics, operational conditions, and virion properties. Among targeted viral surrogates, PMMoV exhibited strongest positive correlation with norovirus GII in membrane permeate samples. Finally, viral infectivity was assessed using the cis-diammine dichloroplatinum (II) (CDDP) capsid integrity test and compared to the performance of the RNase One integrity assay in ultrafiltration feed samples. Preliminary findings suggest that reducing free MNV to below the detection limit requires a CDDP concentration of 5000 µM, which results in approximately a 1-log reduction in intact viruses due to toxicity.