Extending the Life of Wastewater Membranes Significantly Reduces Life Cycle Costs

Abstract
Despite the significant reduction in membrane manufacturing costs, there is a perception within the engineering community that membrane treatment (i.e., MBRs) is more costly than conventional activated sludge systems. Membrane replacement costs have a significant impact on the overall net present value (NPV) of a facility. Membranes installed over the last decade can offer valuable information in estimating membrane life expectancy. Since replacement costs are very expensive, owners are paying more attention to operational adjustments that can extend membrane life. The City of Brandon is one such facility that installed their original membranes over 13 years ago and have observed minimal decline in performance. There is growing evidence that membrane life is determine by operational aspects more than any other factor. Since the City was new to operating an MBR facility in 2009, they began operation of the facility with strict adherence to the manufacturers protocol, and over time have made adjustments which they believe have had favorable impacts on membrane life. This presentation will highlight some of the key considerations made to the operation of the facility.
Background
In October 2009 the City of Brandon began operation of a three train MBR system. In December 2013, the City completed an expansion of the MBR facility, which included the addition of six more membrane trains. At the time the original membrane cassettes were installed in 2009 it was thought that the membrane life would be between 5-7 years. At the end of the 7th year, the membranes were performing very well, and the City decided to continue without replacement. By the end of the 11th year, there was still no decline in performance or increased chemical cleaning requirements. However, due to the uncertainty of how long the membranes would last, the City decided to replace a portion of the 2009 cassettes with new cassettes in December 2020. To get a better idea of how long the membranes will last, the City decided to leave one train (Train 4) populated with original 2009 cassettes. This unique configuration allows the City to operate the facility with differing aged membranes, and use this information to define the replacement strategy for the remaining membranes (see Figure 1).
Objectives
The objective of this paper is to provide a review of the performance of the various aged membranes at the Brandon WRF. The operational adjustments made since start-up will be discussed with a focus on how these adjustments have extended membrane life. Based on the current life expectancy of the membranes the NPV will be compared against the original NPV, providing a better assessment of the cost savings associated with extending the life of the membranes. The NPV information will be provided in the full paper. Membrane Trans-membrane Pressure and Maintenance Cleans Fenu et al (2012) evaluated the concept membrane life and noted that there is little information that defines life expectancy of membranes. Cote et al (2011) suggested that a slow increase in operating pressure and the need for more frequent chemical cleans should be the trigger for replacement. Building on this recommendation, the rate of pressure increase over time between maintenance cleans was evaluated and summarized in Table 1 and Figure 2. The information in Table 1 summarizes the increased pressures following a maintenance clean, done typically around every 60 days. Normal variations in pressure were observed due to varying influent flows, wastewater temperatures, and MLSS filterability. Since all membrane trains were fed the same MLSS and flow, a relative comparison between the pressure in the trains can be made. As expected, the new membranes had lower rate of pressure increase between cleans. However, the 2009 and the 2013 membranes (winter) were identical. During summer months the rate of pressure increase was lower for all trains due to a number of factors (i.e., lower viscosity, lower SRT, etc.).
The information in Table 1 indicates even for the oldest 2009 membranes, the pressure increase is very low allowing the membranes to operate without a maintenance clean for more than 2 months (as compared to the 3 per week originally anticipated). The data indicates that with the 60+ day period between cleans the pressure drop in all trains is restored, indicating an effective removal of foulants. The total number of maintenance cleans conducted over the year were similar for all trains (Table 2). In fact, in 2021 the oldest membrane train (Train 4) received the lowest number of cleans.
Filtered effluent turbidity from each train was similar ranging from 0.05 NTU to 0.08 NTU. In addition to the operating data from the plant, microscopic evaluations of the various aged membranes are currently underway. A preliminary scanning electron microscope image of the 2009 membrane is shown in Figure 3. Details of the microscopy analysis will be provided in the full paper.
Factors Impacting Performance of Brandon's MBRs
The City of Brandon has operated their membrane treatment system for over 13 years, and have made adjustments which they believe extended the membrane life. Operational adjustments discussed in the full paper will include the following: - Proper screening and pre-treatment requirements - Getting the biology right, allowing for colloid adsorption and organic compound degradation. - Reduced maintenance and recovery cleans based on need rather than fixed schedule
Summary
Operational aspects have a significant impact on membrane life, likely more than any other factor. This paper will highlight key adjustments made to the operation of the Brandon MBR facility which have extended their membrane life. Data will be provided on membrane fouling rates, cleaning frequencies, effluent quality, and microscopy analysis. Operational lessons learned will provide other utilities useful information when considering membrane technology, and those assessing the need for membrane replacement. Finally, the NPV of the facility based on the current membrane life will be compared against the original NPV providing an assessment of the cost savings.