J.I. Calvo

Steps of membrane blocking in flux decline during protein microfiltration

The flow decline is studied in typical experiments with dead-end microfiltration of BSA solutions (1 and 0.1 kg/m3) through Cyclopore® track-etched polycarbonate membranes (pore sizes 0.1, 0.2, 0.4 and 1.0 μm) at pH 5 and with a saline content of NaCl 0.01 M. Results are examined, within the frame of the common blocking mechanisms, interpreted here as successive or simultaneous steps of flow decline rather than as alternative possibilities of theoretical modelling for experimental data. An average deposited volume per unit of permeate volume is defined for the central steps of the flux decline, whose behaviour against the shear stress is studied. Under the conditions studied, an increase in shear stress results in a decrease in deposition, probably due to a reduction in protein-surface interaction times. The initial and final steps are also analyzed in terms of an early blockage of the smallest pores and a formation of a cake up to a limiting height, subsequently followed by the establishment of a final non zero flow.

Pore size distributions in microporous membranes. A critical analysis of the bubble point extended method

Abstract An extended bubble point method has been used to examine the porous morphology of several track-etched microporous polycarbonate membranes with nominal pore sizes ranging from 0.1 to 5.0 μm. The technique has been carefully analyzed and corrected to take into account the diverse non-ideal factors in flow along with the prevalence of Knudsen flow over the Hagen-Poiseuille one in the smaller pores. The pore size distributions were obtained in terms of flow or pore numbers. Bulk porosity and pore densities were in good agreement with nominal and SEM values for each filter. Performancewise, measured water permeabilities agreed well with those obtained from pore density distributions, assuming Knudsen air flow through the narrowest pores.

Contact angles and external protein adsorption onto UF membranes

Abstract Both receding and advancing contact angles of pure water and BSA solutions against the active layers of several retentive ultrafiltration membranes are studied here. In particular, three membranes from desalination systems, whose active layers are made of aromatic polyamides: G5, G20 and G50 have been chosen. It has been shown that these membranes are hydrophilic with equal advancing contact angles against pure water while the receding ones decrease with increasing pore radii. Finally, both the contact angles against the BSA solutions are studied at different pH and concentrations. It is shown that the stationary advancing contact angles correspond to the early adsorption steps while receding ones correspond to the final adsorption level. When analyzed within this frame, the results obtained can easily be explained in terms of the qualitatively known molecule–molecule and molecule–membrane surface interactions.

Protein fouling in microfiltration: deposition mechanism as a function of pressure for different pH

The influence of applied pressure on the fouling mechanism during bovine serum albumin (BSA) dead-end microfiltration (MF) has been investigated for a polyethersulfone acidic negatively charged membrane (ICE-450) from Pall Co. BSA solutions at pH values of 4, 5 (almost equal to the protein isoelectric point, IEP), and 6 were microfiltered through the membrane at different applied transmembrane pressures. Results have been analyzed in terms of the usual blocking filtration laws and a substantial change in the fouling mechanism was observed as the pressure was increased, this change can be related to the specific membrane-protein and protein-protein interactions.

Surface structure of microporous membranes by computerized SEM image analysis applied to Anopore filters

The surface morphology of two Anopore® filters (A01 and A02), consisting of thin porous alumina sheets anodically formed, is analyzed here. Scanning Electron Microscopy (SEM) along with Computerized Image Analysis (CIA) have been applied. Both faces of the membranes are studied in order to quantify their presumable asymmetry. The surface porosity and the pore density, or number of pores per surface unit, are directly obtained for each filter side; while, the statistical distributions of the pore areas and perimeters along with equivalent pore diameters and pore shape factors are studied as well. Attending to the found substantial asymmetry, an active membrane side can be distinguished from the porous support. It is seen that support layers are quite similar for both membranes, while the active surfaces have very similar simple pores, being multiple ones which establish differences in mean pore sizes for A01 and A02 membranes. The results on pore size have been used to calculate the water permeabilities of the membranes studied that, when compared with the experimental ones, allow to model the membrane as consisting of an active layer, whose thickness is obtained, plus a porous background.

Mechanisms of protein fouling in cross-flow UF through an asymmetric inorganic membrane

Abstract Flux decline and retention of aqueous solutions of several proteins (pepsin, BSA, lipase, γ-globulin and invertase) with molecular weights of 36, 67, 80, 150 and 270 kDa are studied when they are tangentially filtered through an inorganic microporous membrane with a nominal pore size of 0.02 μm made by Anopore under a constant applied pressure of 5.0 kPa. The fouling process of this ultrafiltration membrane is investigated as a function of the protein concentration, the tangential velocity at the membrane surface and the nature and size of the protein. Finally, the deposition mechanism on the membrane surface and/or into its porous structure is analyzed in terms of several kinetic models. It is seen that an initial particularly intense flux decline due to external blockage is followed by an internal deposition (partially retained proteins) or the formation of a cake (totally retained proteins). The kinetics of these fouling mechanisms are proved to follow reasonable trends versus feed concentrations, recirculation velocity and the protein molecular weight.

Structural characterization of an UF membrane by gas adsorption-desorption and AFM measurements

Abstract In this work, the pore size distributions referred to volume, surface and number of pores, along with the internal surface area and the surface roughness parameters are determined for a capillary microporous symmetric membrane with a nominal diameter of 150 A. This has been realized by using two specially interesting techniques: nitrogen adsorption-desorption at 77 K and atomic force microscopy (AFM). Adsorption isotherms, combined with the BET theory for multilayer adsorption allowed us to obtain the internal surface area of the membrane, while the volume, surface and pore number distributions were calculated from the Kelvin equation both in the adsorption and desorption processes. In this way the existing asymmetry between both adsorption and desorption processes was analyzed and it has been related to the dead-end pores (pores closed to flux). The results of image analysis on AFM micrographs allowed the determination of the surface roughness parameters of the membrane, along with the pore sizes on the membrane surface. Finally, the results have been thoroughly analyzed and compared with other characterization results previously obtained for these membranes from different characterization techniques by other authors.

Bulk and surface characterization of composite UF membranes Atomic force microscopy, gas adsorption-desorption and liquid displacement techniques

Abstract In this work, the pore size distributions referred to volume, surface and number of bulk pores, along with the internal surface area and the size parameters of surface pores are determined for two polyethersulphonic microporous composite membranes of nominal MWCO of 4000 and 30 000 Da. The microporous support pore size distribution has also been obtained. These goals have been achieved using several characterization techniques: atomic force microscopy (AFM), N 2 adsorption-desorption at 77 K and a liquid displacement technique. A computerized analysis of the AFM micrographs allowed determination of the pore size distribution of surface pores at different magnifications. Also, surface roughness can be obtained. Adsorption isotherms, combined with the BET theory for multilayer adsorption, allowed determination of the internal surface area of the membrane, while the volume, surface and pore number distributions were calculated from the Kelvin equation, both in the desorption process to obtain the so-called mesopore distribution. Further analysis, by extending the pore size analysis to pores where the Kelvin equation is not valid, allowed determination of the micropore distribution. Finally, the membrane support has been detached and analyzed by a modified bubble point or liquid displacement technique. Analysis and comparison of all results show that several pore populations (including micro- and mesopores) are present in the membrane with considerable differences between surface pores and bulk pores, the latter being those which should determine permeation. Two further meso- and macropore populations could be assigned to the membrane-support transition or the support itself.

PORE SIZE DISTRIBUTIONS OF TRACK-ETCHED MEMBRANES; COMPARISON OF SURFACE AND BULK POROSITIES

Abstract Here we compare surface and bulk porous morphologies for several track-etched membranes. The experimental techniques include approaches with different proper supplementary pore size ranges. In order to analyze bulk porosities we used: a mercury intrusion apparatus, an extended bubble point method and a gas adsorption–desorption technique which allows to detect pores with decreasing sizes. Referring to the surface structure, scanning electron and atomic force microscopies have been used. Finally, a solute retention method has been used to evaluate pore size distributions from operative conditions of operation. In particular from a comparison of bulk and surface pore size distributions, it is noted that permeability is really determined by the bulk pore size distributions as shown for several track-etched membranes. For these membranes, pore entrance sizes are narrower than those determined by permeability which in turn agree with the bulk pore size distribution. Gas adsorption–desorption results are compared with the atomic force microscopy (AFM) data and solute retention ones for membranes with narrower pores. AFM is especially interesting as far as it is the best technique to analyze the surface roughness. It is worth noting that, among the methods leading to bulk porosity those detecting only thoroughly open pores give always more realistic estimations of the actual permeation performances.

FOULING, STRUCTURE AND CHARGES OF A COMPOSITE INORGANIC MICROFILTRATION MEMBRANE

Here a series of structural parameters of a new composite membrane made by depositing a thin layer of titanium oxide (TiO2) on a cordierite support are measured. An extended bubble point method has been used, along with scanning electron microscopy. Afterwards, the permeate flux decline is analyzed when differently charged proteinic solutions, showing no retention for a clean membrane, are permeated. Permeation experiments are performed at constant concentration and pressure for bovine serum albumin (BSA) (negatively charged at neutral pH) and lysozyme (positively charged at neutral pH) aqueous solutions. It is seen that lysozyme, in spite of its significatively smaller size, fouls the membrane more, giving a faster decline kinetics compared with the bovine serum albumin within the frame of the usual models. The net surface charge of the membrane is obtained by measuring the streaming potential that appears when an electrolytic solution is permeated, whose results are interpreted in terms of a fine capillary model for the transport through the pores that takes into account the electrolyte adsorption. Different streaming potential versus electrolyte concentration behaviour is obtained leading to positive net surface charge density for both the clean and BSA fouled membranes. This positive charge is lower for the fouled than the clean membrane which is in accordance with the negative charge of BSA molecules at neutral pH.