AWWA MTC57567

To evaluate the cleaning efficiency of various protocols, an ultrafiltration membrane pilot skid was operated for several months using settled water produced from the Seine River downstream of Paris in order to simulate membrane fouling conditions that would be expected in a typical water treatment plant. The pilot skid was equipped with micromodules to allow for the comparison of different cleaning procedures given the same state of fouling. Cleaning procedures were evaluated with the objectives of simplicity, rapidity and costeffectiveness. To address these goals, a focus was placed on the use of commodity chemicals such as sodium hypochlorite and caustic soda which are commonly used in water treatment. The specific protocols used were selected according to the quality of the ultrafiltration membrane feedwater (settled water quality) and the level of fouling experienced. Several parameters were studied in order to identify a suitable cleaning procedure. These included the order of chemical addition, the cleaning mode utilized (filtration, circulation, or soaking) and the contact time employed. Additionally, the combined effect of sodium hypochlorite and caustic soda was also evaluated. The efficiency of each cleaning procedure was monitored by measuring the membrane specific flux recovery as well as the evolution of fouling components within the cleaning solution. For analysis of organic matter, ultraviolet (UV) absorption and GC-MS were utilized. Removal of inorganic foulants was followed using ICP analysis of aluminum, manganese, and iron concentrations. Results demonstrated that a cleaning solution containing a mixture of sodium hypochlorite and caustic soda was the most effective cleaning protocol for removal of fouling due to organic matter and aluminum salts. In most cases a simple soaking of the membrane or circulation of the cleaning solution for a one hour time period was sufficient to recover more than 90% of the initial permeability. Includes 7 references, tables, figures.