Public potable water supplies are being stressed by growing population, cyclical droughts, and climate change. One way to sustainably augment the potable water supply is to recycle (reuse) wastewater. This is performed by using a tertiary wastewater treatment facility followed by advanced water purification (AWP) which includes; Reverse Osmosis (RO), UV/advanced oxidation and free chlorine disinfection. RO is the core technology of AWP providing a physical barrier to waterborne pathogens and dissolved constituents. Reliable and safe potable reuse requires continuous demonstration of pathogen removal. Since monitoring pathogens directly is prohibitively time consuming, removal is typically demonstrated using total organic carbon (TOC) and electrical conductivity (EC) as a surrogate. TOC and EC offer a maximum achievable log reduction value (LRV) of approximately 2 and 1.5, respectively. Where under controlled conditions measuring actual pathogens, RO systems provide at least a 4-log reduction. Prior research studies show that strontium, a naturally occurring constituent, has a 3.0 – 3.5 LRV for RO systems. Strontium’s higher LRV allows RO systems to receive more accurate treatment credits and allows increased sensitivity to RO membrane breaches.
This application evaluates the feasibility of demonstrating continuous RO pathogen removal by measuring strontium using Cooper Environmental’s Xact 920, a continuous multi-metals monitor based-on energy dispersive X-ray fluorescence (XRF).
Data presented in this paper was collected over a two-month demonstration in 2021, with the Xact 920 configured to provide continuous 1-hr strontium measurements. The Xact 920 was installed at a pilot AWP facility capable of treating 100,000 gallons per day. The pilot plant was equipped with a 2-stage primary (PRO) and closed circuit secondary (CCRO) RO system which together achieves greater than 90% recovery. One Xact 920 was installed with an automatic switching manifold, allowing switching between four sample locations including; PRO Feed, PRO permeate, CCRO Feed (PRO Brine) and CCRO Permeate. Measuring these four locations allows LRV calculations (Equation 1) for both the primary and secondary RO systems. Figure 1 show the AWP RO treatment process and the Xact 920 sample locations.
LRV=Log(Cf )-Log(Cp) Equation 1
Where:
LRV = log removal value
Cf = strontium RO feed concentration
Cp = strontium RO permeate concentration
The Xact 920’s performance was evaluated based on adequate sensitivity (detection limit), mass balance of the PRO system, and comparing the Xact 920 to U.S. EPA’s Method 200.8, an inductively coupled plasma mass spectrometry (ICP-MS) based method, for total strontium concentration and LRV.
The Xact 920’s 99% confidence level detection limit (DL) for all four streams was calculated. For the two permeate streams standard EPA calculations were performed. However, due to the high native strontium concentrations in the two feed streams, XRF spectral detection limits were calculated and propagated to concentration using the measured preconcentration parameters. Table 1 shows the Xact 920’s DLs along with the average concentration and standard deviation for all four sample steams collected during the study. The closest Xact 920 DL is over four standard deviations from the average concentration. Meaning greater than 99.99% of all strontium measurements will be greater than the Xact 920’s DLs. This demonstrates the Xact 920 has adequate sensitivity for this application.
Strontium mass balance/recovery of the PRO system was calculated using Equation 2 on four separate days. The mass of the strontium on the permeate and brine streams should equal the mass on the feed. The average strontium recovery was 99% ± 3%, demonstrating the Xact 920 strontium measurements are in the expected range.
Sr mass recovery=[(Cp×Fp )+(Cb×Fb )]/((Cf×Ff ) )×100% Equation 2
Where:
Cp = strontium PRO permeate concentration
Fp = PRO permeate flow rate
Cb = strontium PRO brine concentration
Fb = PRO brine flow rate
Cf = strontium PRO feed concentration
Ff = PRO feed flow rate
The Xact 920 measurements were compared to laboratory results using EPA 200.8 for both concentration and LRV. Table 2 shows the average percent difference between the Xact 920 and EPA 200.8 for total strontium concentration, during three sampling campaigns. In addition, one inter-laboratory EPA 200.8 comparison was conducted on all four streams to evaluate the variability of EPA 200.8. The Xact 920 had excellent agreement with EPA 200.8 exhibiting similar percent differences as EPA 200.8 has with itself.
The same data used for percent difference analysis in total strontium concentration was also used to calculate the LRV for the PRO and CCRO systems. Table 3 summarizes the measured LRVs for the Xact 920 and EPA 200.8, while Table 4 shows the percent difference analysis for the LRV. The Xact 920 performance was equal to EPA 200.8, demonstrating the Xact 920 can be used for this application.
Lastly, a membrane breach was simulated on the CCRO system to evaluate if the Xact 920 could detect such an event. Figure 2 shows the Xact 920 and EPA 200.8 total strontium measurements over time during the breach event. The Xact 920 and EPA 200.8 track very closely during normal operation and under a simulated breach.
This application demonstrates the Xact 920 can provide accurate, continuous, near real-time total strontium measurements to demonstrate adequate pathogen removal and high LRV crediting of RO systems.
For more information about the Xact 920 technology, please contact Krag Petterson at kragp@sci-monitoring-cn.com.
Comments are closed