The role of dissolved aluminum in silica chemistry for membrane processes

The use of aluminum sulfate (alum) coagulation prior to reverse osmosis (RO) treatment has been shown to be problematic. Membrane fouling was theorized to occur through soluble aluminum (AP +) reacting with ambient silica (H4SiO 2) to form kaolinite (AI2Si2Os(OH)4) within the RO unit. Chelating agents (citrate at 34 mg/L and ethylenediaminetetraacetic acid [EDTA] at 16 mg/L) were tested for their efficacy in controlling aluminum silicate fouling. The results of bench-scale testing demonstrated that both citrate and EDTA did control aluminum silicate formation, citrate more so than EDTA. Additional aluminum-based fouling was encountered when a commercial, phosphonate-based antiscalant - - used to control barium sulfate scaling - - reacted with the excess aluminum despite the presence of either citrate or EDTA.

Geochemical data developed through this project showed that aluminum silicate scale formation is thermodynamically plausible, with kaolinite and muscovite being the most likely silicate end products. Based on the limited experimental data, citrate and EDTA may effectively act as aluminum sequestering agents that may lead to the prevention of aluminum silicate or hydroxide scaling. Adding a commercial antiscalant did not improve the generic chemicals' ability to control for aluminum silicate fouling, and may be a contributing factor in aluminum-based sealant formation. Three different forms o f aluminumbased foulants were potentially identified during this project: (1) aluminum silicates, (2) aluminum hydroxides, and (3) aluminum phosphates. Further research is needed to confirm the presence o f any one o f these compounds preferably through crystallography or X-ray diffraction spectroscopy. Finally, alternative methods of controlling aluminum, such as alum coagulation at reduced pH or ferric-based coagulation, need to be explored.