WATER RESOURCES RESEARCH GRANT PROPOSAL
Project ID: 2003HI26B
Title: Prevention of Colloidal Fouling in Crossflow Membrane Filtration: Searching for Optimal Operation Conditions: Year 2
Project Type: Research
Focus Categories: Treatment, Solute Transport, Models
Keywords: membrane filtration, colloidal fouling, permeate flux decline
Start Date: 03/01/2004
End Date: 02/28/2005
Federal Funds: $21,694
Non-Federal Matching Funds: $32,307
Congressional District: Hawaii 1st
Albert Sechurl Kim
University of Hawaii at Manoa
Crossflow membrane filtration processes involving reverse osmosis (RO), nanofiltration (NF), ultrafiltration (UF), and microfiltration (MF) have steadily gained importance in environmental engineering separations over the past decade. While RO and NF processes are extensively used for desalination of seawater and brackish water, UF and MF take place of conventional water and wastewater treatment as the alternative, substitutional, or supplementary of the treatment. On the other hand, UF and MF also have been utilized as pretreatment of the desalination processes, in which they remove suspended solid, colloidal particles, macromolecules, bacteria, and viruses from feed suspensions. However, in addition to initial invested cost for installation and operation, membrane fouling often gives rise to extra cost for maintenance and/or (even) replacement of the systems. Specifically during the pretreatment, MF and UF undergo three different kinds of mass-stacking phenomena (of which each provides its own flux decline cause), i.e., concentration polarization, colloidal cake formation, and deposition of (fractal) aggregates. Till date, however, only a few research focused on fundamental understanding of the flux decline behavior in the UF/MF filtrations.
In the first year of this project, the PI focused on development of a model regarding hydraulic resistance due to filtered aggregates which actually provide remarkably less flux decline in a solution of high ionic strength, of which behavior is quite opposite to the conventional understanding of colloidal fouling in UF/MF membrane filtrations. To the best of the PI’s knowledge, the PI’s recent paper submitted to Journal of Membrane Science in Nov. 2003 is the very first model developed for removal of colloidal aggregates on the UF/MF membrane surfaces. As another product in the first year project, the second paper regarding “hydrodynamic radius of a fractal aggregate formed in diffusion-limited cluster aggregation regime” is under preparation with completed theoretical calculations, and will be submitted to the same journal soon.
In the second year, the proposed research will be more focused on statistical mechanics i.e., Monte Carlo, simulations using a Beowulf Linux cluster composed of 16 PCs for distributed parallel computation. The aims are (1) to develop a fundamental statistical mechanical approach to identify the transition point from a liquid-like structure to a solid-like structure of colloidal dispersions due to several physico-chemical and operational conditions in the membrane filtrations, (2) to decide the critical permeate flux under which only concentration polarization is dominant cause of the flux decline before any cake layer forms, (3) to use the particle aggregation phenomena for enhancing the permeate flux if cake formation is inevitable due to the feed water chemistry, and (4) of practical important, to make a simulation-based empirical correlation that can be used by engineers for determining optimum operation conditions in pilot and/or real membrane systems. Therefore, this research will significantly enhance our understanding of fouling problems in the pre-treatment systems of UF/MF, specifically in desalination processes, as well as help engineers/operators develop solutions to effectively prevent the membrane fouling.