|M.Sc Student||Nahir Ran|
|Subject||Modeling of the Fouling Layer on UF Membranes for|
Filtration of Treated Effluents as Function of
|Department||Department of Civil and Environmental Engineering||Supervisor||Professor Carlos Dosoretz|
|Full Thesis text - in Hebrew|
The water crisis in the world brings about the increasing need for water recycling. Among the most efficient available technologies for improving treated effluent quality account membrane separation systems. However, the development of a fouling layer on the membrane surface limits the use of membrane systems for the separation of treated effluents. This fouling layer is composed of materials rejected by the membrane, among them organic and inorganic materials. Within those layers a bacterial film is being build that sticks to the membrane surface and blocks the membrane (biofouling) function as to a decrease in the product flux. Cleaning of this layer can be achieved by back-washings and usage of various chemicals, which shorten the membrane's life and increase highly the maintenance costs.
The main objective of this research is to find a quantitative, semi-empirical correlation by which it will be possible to estimate fouling and biofouling development according to the feedwater quality and the hydraulic conditions (flow velocity, pressure). For this purpose, a transient model has been developed.
The research included two main parts, one experimental and another theoretical.
The experimental part involved the construction of a fourth channel-tubular membrane flow cell system (100 kDa ultrafiltration) operated in commercial wastewater treatment plants and the writing of a control and data storage program. The system was fed with either prefiltered (20 micron) secondary effluents or ultrafiltered (80 kDa) tertiary effluents.
The theoretical part of the research involved the development of a transient model that can describe and predict the phenomenon of fouling development while using data that was gathered during the test-runs. The model is based on the of shear-induced diffusion theory and has been written as a superposition of the inert fouling (development of cake filtration) and biofouling (growth of microorganisms and EPS secretion). An algorithm that solves differential equations by numeric methods was written. An optimization was additionally used to match the model outcomes to the actual field results. Sensitivity analysis and validation to the actual results from field experiments were conducted as well.
The analysis of the results showed that in the case of secondary effluents immediate concentration polarization of suspended solids in the feedwater is the dominant factor. Flux improved with the increase of both Reynolds number and pressure. In tertiary effluents soluble nutrients were found not limiting for biofilm fromation and biofouling was dominant. Flux improved with pressure increase whereas Reynolds number had no effect.