Monitoring MBR Performance For Instability After Intentional Damage

Introduction
Membrane bioreactors provide robust removal of pathogens, as proven through ample studies, and demonstrations; this is attributed to a variety of mechanisms related to biological treatment in addition to the physical exclusion of the membrane itself. Difficulty in monitoring pathogen removal is a challenge in the implementation of potable reuse applications. (WHO 2017) Previous work covers baseline testing and assessment of MBR system performance with breaches in membrane integrity through intentional damage (Katz et al, 2021); this paper will continue forward and address the impact of standard operating procedures that can create instability in compromised systems as demonstrated through intentional damage. This includes conditions such as cleaning and pressure decay testing, where operating procedures may temporarily dislodge solids that have been beneficially plugging fiber breaches. Various tests were conducted under different operating modes and with varying levels of fiber damage. The monitoring of MBR system performance and understanding the impacts on effluent quality is critical to the successful operation of MBR systems in advanced water recycling. To address this, the following are methods of monitoring that were explored:
Turbidity - Turbidity monitoring is the standard for MBR permeate quality and has been adopted as the basis for monitoring in various potable water reuse guidance documents. It is a continuous online measurement that is a strong threshold limit for excellent system performance. (WaterSecure 2017)
#Solids-LRV - A method called 'Solids-LRV' was introduced to estimate the removal of pathogens including protozoa, bacteria and virus. The method is based on the standard method for TSS enumeration with a modification for a low-solids stream. The advantage of this modification is the resolution for TSS can be extended by 3 orders of magnitude thereby allowing the measurement of a high level of removal across the MBR system. The method offers the ability to directly measure the removal of particles across the MBR, is corrected to account for the high volumetric concentration factor in the system, and accounts for all pathogen removal mechanisms due to the selected filter size. (Katz et al., 2018; Katz et al,. 2020)
Pathogens - Pathogen reduction is critical in potable reuse systems and gaining credit for pathogen reduction across unit operations is core to the design of multi-barrier treatment schemes. The pathogens that are typically regulated for reduction in these systems are Protozoa and Virus. As an initial indicator of pathogens, measurements of E. Coli and Total Coliforms were made.
Experimental Testing was done at an MBR pilot unit in Ontario, Canada where raw sewage was screened and fed to the MBR system. Baseline information was gathered on the unit under various conditions, and then membranes were then intentionally damaged with increasingly more significant damage. The impacts on the system effluent were monitored after the damage events and after a variety of stress-events such as maintenance cleaning and pressure decay testing, to assess any instability that these events would create on the self-healing of the fibers. A summary of the experimental conditions and treatment set-up is shown in Table 1.
Results
The baseline data was consistent and no significant differences were seen between the different operating conditions. The average turbidity was 0.04 NTU and Solids-LRV was an average of 4.7 log reduction of solids through the MBR system. Baseline E. Coli log reductions were 7.5 and Total Coliform reductions were 7.4 in operation with relax.
Figure 1 shows the system performance after two maintenance cleans (MC), with 7 fiber cuts occurring in between the two cleans. Figure 2 and 3 both show the system performance following different amounts of fiber damage (15 and 60 cuts, respectively), each followed by a pressure decay test (PDT) and maintenance clean to the system. In each of the tests, maintenance cleaning and pressure decay testing both lead to instability in the membrane performance when the fibers were damaged. This instability of effluent quality is seen in both the increased turbidity and the decreased LRV values (Solids and E.Coli LRV). The amplitude of these spikes was impacted by the amount of damage to the system, with the biggest instability occurring when the damage was highest. The system also took the longest time to recover from stress when the amount of damage was the higher, however in all cases, all of the monitored parameters returned to baseline values within a few hours. Due to the nature of the test, the Solids-LRV values are generally reported on a 24-hr frequency, however the E. Coli was measured and reported at a greater frequency, including samples before the stress condition (fiber cutting, maintenance clean, or pressure decay test), at the beginning and end of the first production cycle after the stress condition and the production cycle 1 hour after the stress condition. This is interesting to see because the first samples after the stress condition show high variability, dropping as low as 2.9 in some cases, but the system quickly recovers from this stress, and returns to high levels of E. Coli removal. This shows direct correlation to the turbidity, which also recovers within a few production cycles.
Conclusions
Turbidity, E. Coli removal, and Solids-LRV all showed a strong correlation to each other across the varying levels of damage, demonstrating that there is promise in Solids-LRV being a complement to turbidity to augment system monitoring. This will provide a stronger indication of pathogen reduction and an additional monitoring point for MBR system performance. Where pathogen removal is a critical parameter for MBR performance, system design, permitting and monitoring should consider operational conditions which increase instability of the system, such as limiting the need to regularly use pressure decay monitoring.