The main objective of this work was to develop a predictive model for nanofiltration (NF) membrane processes. This was accomplished by development of a model which describes the transport of electrolytes in NF membranes in terms of three mechanisms: diffusion, convection and electromigration. The model includes the description of concentration polarisation for binary and more complex mixtures of charged electrolytes. The application and utility of the model were studied by identifying the key characteristics of NF membranes, modelling of a selected process, validation using experimental data and finally using the model for prediction including process optimisation. For one membrane (PES5), atomic force microscopy (AFM) showed that the membrane consisted of discrete pores of nanometre dimensions. Characterisation of the membrane using salts and uncharged solutes showed that it is more appropriate to model the membrane as porous rather than homogenous. The membrane was characterised in terms of the structural parameters: the effective pore radius, <I>r<SUB>p</SUB></I><SUB>, </SUB>and the effective ratio of thickness over porosity, <I>Δx/A<SUB>k</SUB></I>, and an electrical parameter: the effective charge density, <I>X<SUB>d</SUB>. </I>Such characterisation for a further membrane (CA30) was found to be very useful in predicting the process of diafiltration of dye/salt solutions. The prediction required that X<SUB>d</SUB> varied as the salt concentration decreased during processing. The model was then used to predict the processing conditions for the whole diafiltration process which includes the pre/post-concentration phases and the diafiltration phase. Finally, a simplified characterisation method was proposed whereby the membranes that are available in the market are characterised using the information given by the membrane manufacturers. Using the range of parameters obtained, analysis of the model showed that the contributions of all three transport mechanisms are very significant and should not be neglected in any predictive models.
| Identifer | oai:union.ndltd.org:bl.uk/oai:ethos.bl.uk:638212 |
| Date | January 1998 |
| Creators | Mohammad, A. W. |
| Publisher | Swansea University |
| Source Sets | Ethos UK |
| Detected Language | English |
| Type | Electronic Thesis or Dissertation |
Page generated in 0.0016 seconds