A three mechanism model to describe fouling of microfiltration membranes

Mathematical modeling of flux decline during filtration plays an important role in both sizing membrane systems and in the understanding of membrane fouling. Protein fouling is traditionally modeled using one of three classical fouling mechanisms: pore blockage, pore constriction or cake filtration. Here, we have developed a mathematical model to describe flux decline behavior during microfiltration accounting for all three classical fouling mechanisms. Pore constriction was assumed to first reduce the size of internal pores. Pore blockage then occurs at the top of the membrane, preventing further fouling to the interior structure. Finally the foulants at the top of the membrane form a cake, which controls the late stages of the filtration. The model prediction shows excellent agreement with experimental data for 0.25 μm polystyrene microspheres filtered through 0.22 μm Isopore membranes (where pore constriction is expected to be minimal) as well as non-aggregated bovine serum albumin solution through hydrophobic Durapore membranes (where pore constriction is expected to dominate). The effects of different fouling mechanisms on the flux decline were characterized by the ratio of characteristic fouling times of the different mechanisms. In this way the model can provide additional insights into the relative importance of different fouling mechanisms as compared to an analysis by a single mechanism model or by derivative plots, and it can be used to provide important insights into the flux decline characteristics.     

Visualisation of polysaccharide fouling on microporous membrane using different characterisation techniques

Although many studies assessed fouling behaviour in microporous membrane processes like membrane bioreactors (MBRs), in situ or direct observation of the fouling layer has not yet been possible. The observation of the fouling layer resulting from the filtration of model solutions allowed better understanding of MBR fouling intensity and mechanisms. In this study, alginate has been used as a model for polysaccharide (one of the main foulants in MBRs). Three visualisation techniques, confocal laser scanning microscopy (CLSM), environmental scanning electron microscopy (ESEM), and direct observation (DO) have been tested to observe the alginate fouling. The work presented in this paper revealed the advantages and limitations of each technique used for this specific application. Although no coating is required for this technique, ESEM allowed distinct non-destructive observation of clean membrane. However, the lack of structure in the alginate fouling layer limited the use of this technique for fouled membranes. While CLSM requires the use of expensive fluorescent markers, DO appeared as the most promising technique for direct and in situ observation of MBR fouling. DO of alginate/bentonite and alginate/bacteria solutions revealed the creation of a well-structured dual fouling system (bentonite-concentrated layer of 50 μm embedded and covered by a concentration polarisation of alginate greater than 240 μm) on the surface of hollow fibre membrane.     

An approach to membrane fouling characterization by confocal scanning laser microscopy

Confocal scanning laser microscopy (CSLM) is an optical microscopic technique that, among other advantages, can provide high-resolution images from different depths of a three-dimensional object, therefore rendering invasive techniques unnecessary for sample preparation. CSLM in fluorescence mode is a powerful technique in biological applications and in the microscopy of food materials. The main goal of the present study is to develop the appropriate strategies so that CSLM can be used for membrane fouling characterization during the filtration of protein solutions. Single and binary solutions of BSA–fluorescein and ovalbumin–Texas red conjugates were filtered using 0.8 μm polycarbonate membranes. Samples of the membranes at the end of the filtration runs were analyzed by CSLM. A standardized protocol for sample analysis by CSLM was developed and applied in this study. The most significant results show that CSLM can be used to visualize BSA–fluorescein and ovalbumin–Texas red conjugates on top of and inside the membranes, and that they can be distinguished when they jointly foul the membrane. Finally, if the appropriate sectioning is applied a 3D reconstruction of the membrane and the adsorbed/deposited protein can be obtained which give information on the fouling morphology.     

Pd encapsulated and nanopore hollow fiber membranes: Synthesis and permeation studies

New types of supported Pd membranes were developed for high temperature H₂ separation. Sequential combinations of boehmite sol slip casting and film coating, and electroless plating (ELP) steps were designed to synthesize “Pd encapsulated” and “Pd nanopore” membranes supported on -Al₂O₃ hollow fibers. The permeation characteristics (flux, permselectivity) of a series of unaged and aged encapsulated and nanopore membranes with different Pd loadings were compared to those of a conventional 1 μm Pd/4 μm γ-Al₂O₃/-Al₂O₃ hollow fiber membrane. The unaged encapsulated membrane exhibited good performance with ideal H₂/N₂ separation factors of 3000–8000 and H₂ flux 0.4 mol/m² s at 370 °C and a transmembrane pressure gradient of 4 × 10⁵ Pa. The unaged Pd nanopore membranes had a lower initial flux and permselectivity, but exhibited superior performance with extended use (200 h). At the same conditions the unaged 2.6 μm Pd nanopore membrane had a H₂ flux of 0.16 mol/m² s and separation factor of 500 and the unaged 0.6 μm Pd nanopore membrane had a H₂ flux of 0.25 mol/m² s and separation factor of 50. Both nanopore membranes stabilized after 40 h of operation, in contrast to a continued deterioration of the permselectivity for the other membranes. An analysis of the permeation data reveals a combination of Knudsen and convective transport through membrane defects. A phenomenological, qualitative model of the synthesis and resulting structure of the encapsulated and nanopore membranes is presented to explain the permeation results.     

Microfiltration of protein mixtures and the effects of yeast on membrane fouling

The effects of yeast cells on membrane fouling by a protein mixture were studied in dead-end filtration. A 0.2 μm cellulose acetate membrane was used with a 1 g/l protein mixture consisting of equal amounts of bovine serum albumin, lysozyme, and ovalbumin. Yeast cells were used either in suspension or as preformed yeast cakes on top of the membrane. A small concentration of 0.022 g/l yeast cells in suspension enhanced the permeate flux and maintained protein transmission at nearly 100%, compared with a 60% reduction in the protein concentration in the permeate obtained after 3 h for the protein mixture filtered alone. Higher suspended yeast concentrations of 0.043 and 0.18 g/l resulted in lower fluxes and intermediate values for the protein transmission. For the three different thicknesses of preformed yeast cakes studied (0.025, 0.05, and 0.10 cm), the cake with intermediate thickness resulted in protein transmission of nearly 100% and the highest permeate flux. The thinner yeast cake resulted in a lower permeate flux, but it maintained protein transmission at nearly 100%, whereas the thicker cake resulted in a reduction in both permeate flux and protein transmission. The mechanism proposed to explain the results is based on the formation of a secondary membrane by the yeast cells on top of the original membrane. This secondary membrane entraps protein aggregates, which would otherwise cause membrane fouling and reductions in permeate flux and protein transmission.     

Solvent-induced swelling of membranes — Measurements and influence in nanofiltration

This paper describes improvements to an apparatus for in-situ determinations of swelling where a linear inductive probe and electronic column gauge with an overall resolution of 0.1 μm was used for measurements of seven variants of polyacrylonitrile (PAN)/polydimethylsiloxane (PDMS) composite nanofiltration membranes in a range of alkane, aromatic and alcohol solvents. The unswollen membranes incorporated PDMS layers between 1 and 10 μm nominal thickness and were manufactured with a radiation and/or thermal crosslinking step.

The tested membranes exhibited a range of swelling dependent on the degree of crosslinking, the initial PDMS layer thickness and the type of solvent. With no applied pressure the PDMS layer on some radiation cross-linked membranes swelled as much as 170% of the initial thickness whilst other membranes were restricted to a maximum swelling of 80%. When a pressure up to 2000 kPa was applied to a membrane then swelling could be reduced to 20% of the value obtained at zero applied pressure. By vertically stacking up to three membrane samples it was possible to determine the swelling of PDMS layers as thin as 1 μm, although higher imposed pressures rendered some results unreliable as the measurement resolution of the apparatus was approached. The results of the swelling experiments are contrasted with crossflow nanofiltration performance in terms of solvent flux and solute rejection.     

Plasmid DNA processing for gene therapy and vaccination: Studies on the membrane sterilisation filtration step

Membrane filtration through 0.2 μm pores is typically the last operation in the production of pharmaceutical grade plasmid DNA. The membrane sterilisation of purified DNA solutions containing plasmids and bacterial artificial chromosomes (BAC) is investigated in this paper. A linear relationship between total DNA transmission and vector size was observed when filtering through 0.2 μm polyvinylidene difluoride (PVDF) membranes. The percentage of DNA transmission assessed spectrophotometrically varied from 98 to 13% for vector sizes ranging from 6 to 116 kb. There was no significant change in transmission during filtration when controlled flux was increased from 0.1 to 2.3 mL/min cm² or DNA concentration changed from 25 to 100 μg/mL. For vectors ≥20 kb; (i) the level of backbone breakage increased with molecular weight, flux and number of filtration passes; (ii) consecutive filtration experiments indicated that greater DNA loss occurred during the first pass of filtration; and (iii) the use of polyethersulfone (PES) membranes with asymmetrical pores improved DNA transmission and decreased DNA damage. The addition of 150 mM NaCl in the formulation buffer improved filtration transmission by 47 and 11% for the 72 and 116 kb vectors, respectively. Complexation with polyethylenimine (PEI) and a lipid–integrin binding peptide (LI) complex did not improve product transmission.     

Sieve mechanism of microfiltration separation

A theoretical model of dead-end microfiltration (MF) of dilute suspensions is proposed. The model is based on a sieve mechanism of MF and takes into account the probability of membrane pore blocking during MF of dilute colloidal suspensions. An integro-differential equation (IDE) that includes both the membrane pore size and the particle size distributions is deduced. According to the suggested model a similarity property is applicable, which allows one to predict the flux through the membrane as a function of time for any pressure, and dilute concentration, based on one experiment at a single pressure and concentration. The suggested model includes only one fitting parameter, β>1, which takes into account the range of the hydrodynamic influence of a single pore. For a narrow pore size distribution in which one pore diameter predominates (track-etched membranes), the IDE is solved analytically and the derived equation is in good agreement with the measurements on different track-etched membranes. A simple approximate solution of the IDE is derived and that approximate solution, as well as the similarity principal of MF processes, is in good agreement with measurements using a commercial Teflon microfiltration membrane. The theory was further developed to take into account the presence of multiple pores (double, triple and so on pores) on a track-etched membrane surface.

A series of new dead-end filtration experiments are compared with the proposed initial and modified pore blocking models. The challenge suspension used was nearly monodispersed suspension of latex particles of 0.45 μm filtered on a track-etched membrane with similar sized pores 0.4 μm. The filtered suspension concentration ranged from 0.00006 to 0.01% (w/w) and the cross-membrane pressures varied from 1000 to 20,000 Pa. Three stages of microfiltration have been observed. The initial stage is well described by the proposed pore blocking model. The model required only a single parameter that was found to fit all the data under different experimental operational conditions. The second stage corresponds to the transition from the blocking mechanism to the third stage, which is cake filtration. The latter stage occurred after approximately 10–12 particle layers were deposited (mass = 0.006 g) on the surface of the microfiltration membrane.     

A microfluidic membrane chip for in situ fouling characterization

A new method for non-invasive in situ monitoring of a microfiltration process is described. In microfiltration systems, local information on the deposition characteristics can be used to determine the cake behavior during a filtration run. Typically, non-invasive methods of fouling study are restricted to specialized membranes, or require highly complex systems. This study employs the use of synthetic embedded channel membranes, with channels separated by a porous structure (active membrane). The characteristics of the active membrane have been analyzed. Deposition on the membrane surface can be observed and monitored optically across the width of the feed channel. This can be used to observe the liquid hydrodynamics in the channel as well as the local cake properties in time. In dead end filtration, it has been observed that with 6 μm particles, the cake initially deposits towards the end of the membrane. However, as filtration continues, the deposition changes with more local deposition towards the channel entrance, leading to a more homogeneous cake layer.     

Crossflow filtration of washed and unwashed yeast suspensions at constant shear under nominally sub-critical conditions

Operation at sub-critical fluxes can be used to control membrane fouling. The original definition of the critical flux stated that operation was sub-critical if no or negligible fouling occurred. Over time there has been a relaxing of the criteria and many now consider a low rather than zero rate of fouling to be indicative of sub-critical operation especially when dealing with complex feeds. Here the region of low fouling is termed “nominally sub-critical”. Unwashed yeast, washed yeast and extra polymeric substance (EPS) suspensions were filtered at controlled fluxes to investigate the role of cells and soluble components in nominally sub-critical conditions using ultrafiltration (UF) and microfiltration (MF) membranes. As the UF membrane could not be effectively cleaned it was not used in the later part of the study. Tracking of membrane resistance, of the 0.2 μm membranes was continued through the whole study. After the initial increase, rises very slowly, increasing on average only 0.4% after each experiment and cleaning cycle. For the MF membranes, the rate of fouling increased with increasing feed concentration, increasing membrane pore size and decreasing shear stress. The effect of increasing shear stress was to reduce the amount of reversible fouling but the irreversible component was invariant with shear stress for the range studied. Also the rate and reversibility of fouling were found to be sensitive to changes in pH. The sum of the rates of transmembrane pressure (TMP) rise for washed yeast cells and EPS suspensions were in all cases found to be lower than that for unwashed yeast. The origin of the additional resistance is discussed and other relevant literature reviewed.     

Effects of vascular endothelial growth factor (VEGF) and chondroitin sulfate A on human monocytic THP-1 cell migration

Angiogenesis serves as a crucial factor in disease development and progression, such as cancer metastasis, and monocyte migration is one of the key steps for angiogenesis. Therapeutic modulation of angiogenesis is a promising new therapeutic avenue under investigation. In this study, effects of vascular endothelial growth factor (VEGF) and chondroitin sulfate A on monocyte migration were investigated. Human monocytic THP-1 cells were from Riken Cell Bank (Tsukuba, Japan) and vascular endothelial cells (VECs) were obtained from swine thoracic aorta. The migration experimental system was adapted from Falcon™ Cell Culture Inserts with pore sizes of 3 and 8 μm cultured endothelial cells or not on the insert polyethylene terephthalate (PET) membranes. Four VEGF concentrations (0, 10, 50 and 100 ng/ml) and three concentrations of chondroitin sulfate A (0, 1.25 and 5.0 mg/ml) were used to investigate their effects on THP-1 cell migration ability through PET membranes and VECs monolayer. The THP-1 cell migration was evaluated by counting the number of migrated cells related to the total number of cells under a microscope. We counted the migration cells every 1 h on a Tatai-type hemocytometer using an inverted microscope for total 7 h. For inserts with pore sizes of 3 and 8 μm, the THP-1 cell migration increased with VEGF concentrations; however, cell migration decreased with the chondroitin sulfate A concentration. Our results demonstrated that VEGF accelerated monocyte migration through endothelial monolayer and chondroitin sulfate A is an effective inhibitor of monocyte migration for angiogenesis.     

An NMR study of methanol diffusion in polymer electrolyte fuel cell membranes

Methanol diffusion in two polymer electrolyte membranes, Nafion 117 and BPSH 40 (a 40% disulfonated wholly aromatic polyarylene ether sulfone), was measured using a modified pulsed field gradient NMR method. This method allowed for the diffusion coefficient of methanol within the membrane to be determined while immersed in a methanol solution of known concentration. A second set of gradient pulses suppressed the signal from the solvent in solution, thus allowing the methanol within the membrane to be monitored unambiguously. Over a methanol concentration range of 0.5–8 M, methanol diffusion coefficients in Nafion 117 were found to increase from 2.9 × 10⁻⁶ to 4.0 × 10⁻⁶ cm² s⁻¹. For BPSH 40, the diffusion coefficient dropped significantly over the same concentration range, from 7.7 × 10⁻⁶ to 2.5 × 10⁻⁶cm² s⁻¹. The difference in diffusion behavior is largely related to the amount of solvent sorbed by the membranes. Increasing the methanol concentration results in an increase in solvent uptake for Nafion 117, while BPSH 40 actually excludes the solvent at higher concentrations. In contrast, diffusion of methanol measured via permeability measurements (assuming a partition coefficient of 1) was lower (1.3 × 10⁻⁶ and 6.4 × 10⁻⁷ cm² s⁻¹ for Nafion 117 and BPSH 40 respectively) and showed no concentration dependence. The differences observed between the two techniques are related to the length scale over which diffusion is monitored and the partition coefficient, or solubility, of methanol in the membranes as a function of concentration. For the permeability measurements, this length is equal to the thickness of the membrane (178 and 132 μm for Nafion 117 and BPSH 40 respectively) whereas the NMR method observes diffusion over a length of approximately 4–8 μm. Regardless of the measurement technique, BPSH 40 is a greater barrier to methanol permeability at high methanol concentrations.     

Titania membrane preparation with chemical stability for very hash environments applications

The high-quality tubular titania MF membranes are successfully prepared by dip-coating techniques and systematically investigated with regard to their corrosive resistances. The experiments show that dispersants PAA and anatase powder were preferably employed to prepare desired suspensions with solid loading 10–15 wt.% and that suspensions properties significantly affect the final membranes quality. The titania MF membranes with pure water permeability 742.42 l m⁻² h⁻¹ bar⁻¹ and 0.1 μm pore diameter have been obtained using the prepared suspensions. The corrosion resistance of titania membranes exhibits that the anatase layers are more stable than the alumina supports in boiling corrosive medium and that the poor quality titania MF membranes become more deteriorated due to its top layer pore blockages and fouling layer formation, which provides a wide range of practical application fields in very hash environments with reliable data supports.     

Preparation of P(AN–MMA) microporous membrane for Li-ion batteries by phase inversion

A novel process was proposed for preparation of microporous poly(acrylonitrile–methyl methacrylate) (P(AN–MMA)) membranes by phase inversion techniques using ultrasonic humidifier. Being prepared by dissolving the polymer (PAN–MMA) in the N,N-dimethylformamide (DMF) solution with mechanical stirring, the homogenous casting solution was cast onto a clean glass plate. Successively, the glass plate was exposed to the water vapor produced by ultrasonic humidifier, inducing the phase inversion. It is found the pore size is much more uniform across the cross-section of the membrane than that of the porous membrane prepared by conventional water bath coagulation technique. The microporous membranes were directly obtained after the washing and drying. It had about 1–5 μm of pores and presented an ionic conductivity of 2.52 × 10⁻³ S/cm at room temperature when gelled with 1 M LiPF₆/EC-DMC (1:1 vol.%) electrolyte solution. The test cells with the gel electrolytes prepared from as-prepared microporous membranes showed stable cycling capacities, indicating that the microporous membrane, which was prepared from cheap starting materials acrylonitrile and methyl methacrylate, can be used for the gel electrolyte of lithium batteries.     

Characterization of fibril reinforced membranes for fuel cells

Characteristics of fibril reinforced membranes developed by Asahi Glass Company are reviewed. PTFE-fibrils <1 μm in diameter are dispersed in ion-exchange membranes uniformly. Mechanical properties, such as tensile strength, tear strength, creep property and compressive property were examined and compared with non-reinforced membranes. Fibril reinforced membranes, even by the addition of a small amount of PTFE-fibrils (2.7 wt.%), show excellent mechanical strength, especially in creep and tear strength. Cell performance is nearly equal to the one using a non-reinforced membrane and cell voltage stays about the same during the cell operation at 80 °C for 3000 h.     

Solar driven membrane pervaporation for desalination processes

We describe details of a solar driven pervaporation process for the production of desalinated water from highly contaminated waters. The membrane material is a polyetheramide-based polymer film of 40 μm thickness. This Solar Dew^® membrane is used in a tubular configuration in a direct solar membrane pervaporation process. The feed waters used in this study are untreated seawater and waste water that is simultaneously produced with the mineral oil extraction. In all cases retention of typical ions as sodium, chloride and calcium as well as specific problematic ions (arsenic, boron and fluoride) was higher than data reported for pressure driven membrane processes like NF and RO. The condensate quality was well within WHO limits for drinking water. A reduction of almost five orders of magnitude in conductivity between brine and condensate could be realized, producing condensate with conductivities of 5 μS/cm or lower. Laboratory experiments show that the measured fluxes are independent of severe fouling and virtually independent of concentration up to 100 g/l total solids.     

Blended chitosan and polyvinyl alcohol membranes for the pervaporation dehydration of isopropanol

Homogeneous membranes were prepared by casting the solution of blended chitosan and polyvinyl alcohol (PVA) on a glass plate. The percent weight of chitosan in the membrane was varied from 0 to 100%. The membrane thickness was in the range of 15–30 μm. The membranes were heat treated at 150 °C for an hour. After that the membranes were crosslinked by glutaraldehyde and sulfuric acid in acetone aqueous solution. The membranes were tested at 30–60 °C for dehydration performance of 50–95% isopropanol aqueous solutions. At around 90% of isopropanol in the feed mixture, permeate flux increased whereas the percent of water in permeate tended to decrease when the feed temperature increased for all membranes, except that the water content in permeate from the membrane containing 75 wt.% chitosan remained constant. The swelling degree in water and the total flux increased with increasing chitosan content in membranes. The effect of temperature on permeate flux followed the Arrhenius relationship. The permeate flux decreased when isopropanol in the feed increased for all membranes. However, water content in permeate and isopropanol concentration in the feed formed complex relationship for different chitosan content membranes. Sorption did not appear to have significant effects on separation. The membrane containing chitosan 75% performed the best. For a feed solution containing 90% isopropanol at 60 °C, the permeate flux was 644 g/m² h with water content of nearly 100% in the permeate. At 55% isopropanol in the feed at 60 °C, the permeate flux was 3812 g/m² h. In the range of 55–95% of isopropanol in the feed, the water content in permeate was more than 99.5%. This membrane showed very excellent performance with good mechanical strength. It is promising to develop this membrane for industrial uses.     

Effect of laser modified surface microtopochemistry on endothelial cell growth

The introduction of microelectronics technology in the area of biological sciences has brought forth previously unforeseeable applications such as DNA or protein biochips, miniaturized, multiparametric biosensors for high performance multianalyte assays, DNA sequencing, biocomputers, and substrates for controlled cell growth (i.e. tissue engineering). We developed and investigated a new method using “cold” excimer laser beam technology combined with microlithographical techniques to create surfaces with well defined 3D microdomains in order to delineate critical microscopic surface features governing cell–material interactions. Microfabricated surfaces with microgrooves 30–3 μm deep, 10–1 μm wide spaced 30 μm apart were obtained with micron resolution, by “microsculpturing” polymer model surfaces using a computer controlled laser KrF excimer beam coupled with a microlithographic projection technique. The laser beam after exiting a mask was focused onto the polymer target surface via an optical setup allowing for a 10-fold reduction of the mask pattern. Various 3D micropatterned features were obtained at the micron level. Reproducible submicron features could also be obtained using this method. Subsequently, model human umbilical endothelial cells (HUVEC) were cultured on the laser microfabricated surfaces in order to study the effects of specific microscopic surface features on cell deposition and orientation. Cell deposition patterns were found to be microstructure dependant, and showed cell orientation dependency for features in the cell range dimension, a behaviour significantly different from that of a previously studied cell model (osteoprogenitor cell). This model may be a promising in so far as it is very rapid (a time frame less than a second per square centimeter of micropatterned surface) and provides further insights into the effects of surface microtopography on cell response with possible applications in the field of biosensors, biomedical and/or pharmaceutical engineering sciences.     

Unsupported palladium alloy foil membranes fabricated by electroless plating

It is desirable to create thin (<25 μm), unsupported, defect-free palladium and palladium alloy foils in a cost-effective manner in order to study intrinsic material properties exclusive of support effects. We have developed a novel technique for producing unsupported palladium films by electroless plating upon mirror-finished stainless steel supports followed by mechanical removal. High quality pure palladium films as thin as 7.2 μm were produced. Single gas steady state permeation experiments were performed using hydrogen and nitrogen to examine permeability and selectivity. The pure palladium membranes showed hydrogen permeabilities comparable to cold-rolled unsupported foils, and high H₂/N₂ selectivity. Palladium-copper membranes were prepared by sequential electroless plating of copper onto palladium foils followed by in situ annealing. The annealing process produces films of desired composition with permeabilities comparable to those in the literature. The annealing process does not appear to produce defects in the film, and the membranes thus produced have performed 15 days without increased leak rates.     

Polysilicato-iron for improved NOM removal and membrane performance

The natural organic matter (NOM) removal efficiency of polysilicato-iron (PSI) coagulants and the fouling potential of PSI pretreated waters have been studied using two microfiltration (MF) membrane types: polyvinylidene fluoride (PVDF-2) and polypropylene (PP). The results showed that PSI coagulant with a Si/Fe ratio of 1 (PSI-1) was the most effective, compared to conventional coagulants, in removing dissolved organic carbon (DOC) and in improving the fouling potential. A relative flux of unity through PVDF-2 membrane was achieved for both water sources pretreated with PSI-1.

Aluminium-based coagulants, particularly aluminium chlorohydrate (ACH), worked best at lower coagulant dose. Increasing the coagulant dose to improve DOC removal led to increased membrane fouling, possibly due to increased level of unsettleable flocs and pore blocking. For PSI with larger floc size, the advantage of increased DOC removal was not overridden by the adverse effect of pore blocking. In addition, the residual neutral fraction in the waters and/or the presence of a filter cake on the membrane surfaces seemed to have a limiting effect on the fouling rates through both PP and PVDF-2 membranes to the extent that similar rates were obtained, despite substantial differences in DOC removal.

In contrast, these limiting factors did not influence the fouling potential of PSI-1 treated waters through the PVDF-2 membrane, as suggested by the relative flux of unity for both water sources. It is suggested that the oxide deposits on the PVDF-2 membrane may act as a ‘screening layer’, acting as pre-filtration by the filter cake. This layer may be effectively removed by backwashing, together with deposited NOM, throughout the experiment to maintain the flux at unity. The hydrophobic nature of the PP membrane may discourage the deposition of the oxides, thus minimising the positive effects of the oxides in the system. The high removal of hydrophobic fractions by PSI-1 may also lead to less association between residual NOM and less binding to the membranes, particularly on the PVDF-2 membrane.