M.H.V. Mulder

Retention measurements of nanofiltration membranes with electrolyte solutions

Retention measurements with single salt solutions of CaCl2, NaCl and Na2SO4 revealed that the rejection mechanism of commercial polymeric nanofiltration membranes investigated in this study may be divided into two categories: 1. Membranes for which Donnan exclusion seems to play an important role. 2. Membranes for which retention is determined by both Donnan exclusion and size effects. In category 1 both positively and negatively charged membranes were found. Ceramic γ-Al2O3 ultrafiltration membranes with a pore size of 3 nm showed a same type of salt retention behavior as the positively charged polymeric membranes. The extended Nernst–Planck equation in combination with the Donnan equilibrium has been used to model the flux-retention experiments for the salt solutions. The numerical calculations resulted in a good agreement with experimental data and acceptable values for the fixed charge densities have been determined. The effective membrane thicknesses calculated were higher than those observed by scanning electron microscopy.

Zeolite-filled silicone rubber membranes : Part 1. Membrane preparation and pervaporation results

Amongst the alternative fuels obtained from renewable resources alcohol from fermentation may become one of the most important. The combination of fermentation with pervaporation in a membrane bioreactor offers the advantage of continuous processing. In this membrane bioreactor alcohol-selective membranes are needed. The performance of the membranes available at present is poor. Much research is being carried out on silicone rubber but the selectivity of this material for alcohol is too low. Addition to the membrane of a sorptive filler with a high selectivity towards alcohol appears to improve both selectivity and flux. Silicalite, a novel type of hydrophobic zeolite, has been used for that purpose. Results presented in this paper indicate that transport through the zeolite pores contributes to a major extent to the total transport through the membrane.

Characterization of polymer blends of polyethersulfone/sulfonated polysulfone and polyethersulfone/sulfonated polyetheretherketone for direct methanol fuel cell applications

Existing polymer electrolyte membranes (PEMs) applied for hydrogen fuel cells are frequently not suitable for direct methanol fuel cells due to the high methanol permeability. Therefore, new materials are required and in order to avoid laborious fuel cell experiments with a so-called membrane–electrode assembly (MEA) proper characterization methods are needed which can be used as a first estimate. Various methods have been selected, such as swelling measurements, ion-exchange capacity (IEC), electrical resistance, permselectivity, sorption isotherms and methanol permeability which give information on the conductive properties and methanol permeability properties of the materials. These methods have been applied on blends of polyethersulfone (PES)/sulfonated polysulfone (SPSf) and PES/sulfonated polyetheretherketone (SPEEK) in which the composition has been changed to obtain defined properties. Based on the results, it can be concluded that the methods can be used as a first estimate to select proper materials for direct methanol fuel cell applications.

Wetting criteria for the applicability of membrane distillation

Membrane distillation can only be applied on liquid mixtures which do not wet a microporous hydrophobic membrane. Solutions of inorganic material in water have such high values of surface tension (γLgreater-or-equal, slanted72x10−3 N/m) that the non-wetting condition is fulfilled for a number of hydrophobic membranes. As soon as organic solutes are present in the solution, the surface tensionγL will be lowered, and if the concentration of organic material becomes too high, wetting of the membrane will occur. By means of theoretical considerations a critical solute concentration or surface tension at which a homogeneous smooth material will be wetted (gq < 90/deg) can be calculated. For a (micro)porous membranes no such theoretical relation can be derived. Therefore, a simple experimental method is described to measure the maximum allowable concentration for a (micro)porous membrane. On the basis of these measurements, the maximum allowable concentration under process conditions can be determined.

Adsorbent filled membranes for gas separation. Part 1. Improvement of the gas separation properties of polymeric membranes by incorporation of microporous adsorbents

The effect of the introduction of specific adsorbents on the gas separation properties of polymeric membranes has been studied. For this purpose both carbon molecular sieves and zeolites are considered. The results show that zeolites such as silicate-1, 13X and KY improve to a large extent the separation properties of poorly selective rubbery polymers towards a mixture of carbon dioxide/methane. Some of the filled rubbery polymers achieve intrinsic separation properties comparable to cellulose acetate, polysulfone or polyethersulfone. However, zeolite 5A leads to a decrease in permeability and an unchanged selectivity. This is due to the impermeable character of these particles, i.e. carbon dioxide molecules cannot diffuse through the porous structure under the conditions applied. Using silicate-1 also results in an improvement of the oxygen/nitrogen separation properties which is mainly due to a kinetic effect. Carbon molecular sieves do not improve the separation performances or only to a very small extent. This is caused by a mainly dead-end (not interconnected) porous structure which is inherent to their manufacturing process.

Vapour sorption and permeation properties of poly (dimethylsiloxane) films

Sorption and permeation of several organic vapours in poly (dimethylsiloxane) were investigated. Solubility and permeability coefficients measured show a strong dependence of the applied vapour activity typical for these systems, contrary to the behaviour of permanent gases. However, as irregularities in the series of chloromethanes show, highest sorption does not necessarily lead to the highest permeability value. A possible explanation of this effect can be found in the concentration dependent diffusion behaviour of poly (dimethylsiloxane) toward these organic vapours. Thus, from the kinetics of solvent uptake, i.e. the sorption isotherms over time, the diffusion coefficients, which were corrected for swelling and thermodynamic activity of the vapour, were calculated for the different vapour activities.

Preparation of composite hollow fiber membranes: co-extrusion of hydrophilic coatings onto porous hydrophobic support structures

Coating a layer onto a support membrane can serve as a means of surface functionalization of membranes. Frequently, this procedure is a two-step process. In this paper, we describe a concept of membrane preparation in which a coating layer forms in situ onto a support membrane in one step by a co-extrusion process. Our aim is to apply a thin ion exchange layer (sulfonated polyethersulfone, SPES) onto a polysulfone support. The mechanical stability and adhesion of the ion-exchange material to the hydrophobic support membrane (polysulfone) has been studied by a systematic approach of initial proof-of-principle experiments, followed by single layer and double-layer flat sheet casting. Critical parameters quantified by the latter experiments are translated into the co-extrusion spinning process. The composite hollow fiber membrane has low flux as a supported liquid membrane for the copper removal due to the low ion exchange capacity of the SPES. The coating layer of the composite membrane is porous as indicated by gas pair selectivity close to unity. However, our new composite membrane has good nanofiltration properties: it passes mono and bivalent inorganic salts but rejects larger charged organic molecules. The experimental work demonstrates that co-extrusion can be a viable process to continuously prepare surface tailored hollow fiber membranes in a one-step process, even if the support and coating material differ significantly in hydrophilicity.

ON THE MECHANISM OF SEPARATION OF ETHANOL/WATER MIXTURES BY PERVAPORATION II. EXPERIMENTAL CONCENTRATION PROFILES

Ethanol—water concentration profiles in cellulose acetate membranes were measured under steady-state pervaporation conditions. Knowledge of these profiles leads to a better understanding of the diffusion process during pervaporation. The concentration profiles were determined by a film-stack method, using three to six layers. It is shown that permeation of ethanol—water mixtures proceeds in a coupled way and that crossterm diffusion coefficients need to be considered. Furthermore, the occurrence of sorption resistances at the feed/membrane interface can be established from these experiments

Pervaporation of alcohol-toluene mixtures through polymer blend membranes of poly(acrylic acid) and poly(vinyl alcohol)

Homogeneous membranes were prepared by blending poly(acrylic acid) with poly(vinyl alcohol). These blend membranes were evaluated for the selective separation of alcohols from toluene by pervaporation. The flux and selectivity of the membranes were determined both as a function of the blend composition and of the feed mixture composition. The results showed that a polymer blending method could be very useful to develop new membranes with improved permselectivity. The pervaporation properties could be optimized by adjusting the blend composition. All the blend membranes tested showed a decrease in flux with increasing poly(vinyl alcohol) content for both methanol?toluene and ethanol?toluene liquid mixtures. The alcohols permeated preferentially through all tested blend membranes, and the selectivity values increased with increasing poly(vinyl alcohol) content. The pervaporation characteristics of the blend membranes were also strongly influenced by the feed mixture composition. The fluxes increased exponentially with increasing alcohol concentration in the feed mixtures, whereas the selectivities decreased for both liquid mixtures.

Removal of trace organics from aqueous solutions. Effect of membrane thickness

A resistance-in-series model is used to describe the pervaporation performance of elastomeric membranes in the removal of volatile organic components from water. Equations have been derived to describe the organic component flux as a function of feed concentration, permeability of the organic component in the membrane, membrane thickness and liquid boundary layer mass transfer coefficient. The model has been verified using both homogeneous and composite membranes of polydimethylsiloxane, ethylene propylene rubber and polyoctenamer. Membranes with a wide range of thicknesses have been prepared and the pervaporation behaviour for the removal of toluene and trichloroethylene from aqueous solutions has been studied. The experiments show that the hydrodynamic boundary layer resistance is of great importance. For highly permeable polymers such as polydimethylsiloxane mass transfer in the boundary layer is rate determining and should be considered carefully in further development of the process. For less permeable polymers such as ethylene propylene rubber this effect becomes more dominant with decreasing membrane thickness. The water fluxes are inversely proportional to the thickness of the actual separating layer and they depend strongly on the type of elastomer used. A proper choice of the elastomeric material and the thickness of the separating layer will determine the selectivity of the process.