Scalability and Design of a Submerged Membrane Bioreactor for Municipal Wastewater Treatment

The topic of membrane performance and fouling in a membrane bioreactor system (MBR) has been extensively researched. Many approaches have been employed to explore the mechanisms behind fouling and how to control it. While a very well-studied topic, the vast majority of published results are based on lab-scale or small pilot-scale systems using synthetic wastewater. Very few studies have looked at the ability to scale the results of these studies to apply at full-scale. Use of real wastewater and testing at larger scales has been extremely limited. In an effort to close this fundamental research gap, a multi-year study on the impact system scale has on the short term membrane performance and fouling in MBR systems was conducted. Four systems of increasing size from a lab-scale system to a full-scale 3.0 million gallon/day plant were used to conduct the study. All four systems operated with municipal wastewater and used the same hollow fibre membrane. The applicability of scaling the results of small systems was explored by introducing a new evaluation approach using the difference in the ratio of solid mass flux (SMF) handled by the membranes to the amount of air mass flux (AMF) introduced to control membrane fouling. Recommendations on how to use small system data were proposed based on the findings. The impact key membrane module operational and design parameters have on fouling were also examined. The novel SMF to AMF analysis was applied to allow for comparison of results and helped determine which changes could lead to the best performance optimization. Finally, an Excel® based model capable of predicting lifecycle costs for MBR systems was developed. The model was found to be useful in predicting costs for up to six membrane products simultaneously. The impact of membrane design and operational parameters on lifecycle costs was explored using three similar products. The analysis included the impact of key operational and design parameters over a wide range of plant flow capacities. Results show that product properties and operational parameters can have a larger than expected impact on the lifecycle costs of a system.