An enzymatic approach to the cleaning of ultrafiltration membranes fouled in abattoir effluent

Membrane fouling severely curtails the economical and practical implementation for the purification of biologically related process streams such as abattoir effluent (Jacobs, WRC Report no. K5/362, 1991, Pretoria, South Africa [1]). Mechanical and chemical removal of foulants usually lead to membrane damage and additional pollution. Enzymes, specific for the degradation of proteins and lipids, were tested as key components of biological cleaning regimes for membranes fouled in abattoir effluent. Fouling of polysulphone membranes was assessed as previously described by Maartens et al. (J. Membrane Sci., 119 (1996) 1 [2]) and optimal enzyme concentrations and incubation times were determined for the different preparations. The ability of each cleaning agent to remove adsorbed protein and lipid material, as well as their ability to restore the water-contact angle and the pure-water flux of the fouled membranes, were determined and compared. These variables were also used to compare the cleaning efficiency of enzymatic cleaning agents with conventional chemical agents under optimal conditions. The enzymes and enzyme detergent mixtures were effective cleaning agents and the pure-water flux of statically fouled membranes could be restored by treatment with these agents.     

Predicting the effect of membrane spacers on mass transfer

Like many separation processes, ultrafiltration and reverse osmosis are often compromised by concentration polarization. Such polarization can be mitigated by static mixers and other flow barriers placed as spacers next to the membrane surface. These spacers can be shaped like ladders, herringbones, and helices. The effect of these spacers can be successfully predicted without adjustable parameters from extensions of the Lévêque equation. The predictions are in agreement with results of computational fluid mechanics and with electrochemical experiments. They supply a tool for optimizing spacer design.     

Enzymatic cleaning of ultrafiltration membranes fouled by protein mixture solutions

Compared to other typical cleaning agents, application of enzyme in cleaning of membranes fouled with protein solution promised the high cleaning efficiencies with lower environmental impact. This paper is focused on the mechanisms of protein removal by enzyme cleaning agent from the membrane surface by analysis hydraulic resistance, total protein removal using Lowry method, and membrane surface analysis using MALDI-MS and gel electrophoresis to estimate the foulant composition. Using single and binary protein solutions of bovine serum albumin and beta-lactoglobulin as the feed solution for filtration process, the experimental results indicate that optimum cleaning time and cleaning agent concentration is due to the competition between foulant removal and deposition of enzymes on the membrane during the cleaning process. The removal rate of different protein species in the fouling layer is varied, indicating that cleaning strategies can be tailor-made for fouling layer with different protein compositions.     

Chiral separation of phenylalanine in ultrafiltration through DNA-immobilized chitosan membranes

Ultrafiltration experiments for the chiral separation of racemic phenylalanine were performed with DNA-immobilized chitosan membranes having various pore sizes. Atomic analysis on the membranes showed that the chitosan membranes covalently bound six times more DNA than the cellulose membranes used in our previous study [A. Higuchi, Y. Higuchi, K. Furuta, B.O. Yoon, M. Hara, S. Maniwa, M. Saitoh, K. Sanui, Chiral separation of phenylalanine by ultrafiltration through immobilized DNA membranes, J. Membr. Sci. 221 (2003) 207–218]. d-Phenylalanine preferentially permeated through DNA-immobilized chitosan membranes with a pore size <6.4 nm [molecular weight cut-off (MWCO) <67,000]. The binding affinity of a specific enantiomer due to the pore size of the DNA-immobilized membranes regulated the preferential permeation of the enantiomer through the membranes. l-Phenylalanine was adsorbed on the DNA-immobilized chitosan membranes with a pore size <6.4 nm (MWCO < 67,000), while there was little difference between the adsorption of d-phenylalanine and l-phenylalanine on the membranes with a pore size >6.4 nm (MWCO > 67,000). The DNA-immobilized chitosan membranes were categorized as channel type membranes.     

Experimental study of ultrafiltration membrane fouling by sodium alginate and flux recovery by backwashing

Membrane fouling is the major limitation for a broader application of membrane technology. One of the main causes of membrane fouling in advanced wastewater reclamation and in membrane bioreactors (MBR) are the extracellular polymeric substances (EPS). Among the main constituents in EPS, polysaccharides are the most ubiquitous. This study aims at a better understanding of the fouling mechanisms of EPS and the efficiency of backwashing technique, which is applied in practice to restore membrane flux. For that purpose, the evolution of fouling by sodium alginate, a microbial polysaccharide, is studied in ultrafiltration. Fouling experiments are carried out in a single fiber apparatus, aiming at identifying the significance of distinct fouling mechanisms and their degree of reversibility by backwashing. An important parameter considered in the study is the concentration of calcium ions, which promote sodium alginate aggregation and influence the rate of flux decline, the reversibility of fouling and rejection. A rapid irreversible fouling takes place due to internal pore constriction, at the beginning of filtration, followed by cake development on the membrane surface. With increased calcium addition, cake development becomes the dominant mechanism throughout the filtration step. Furthermore, fouling reversibility is increased with the increase of calcium concentration. A unique behavior of sodium alginate solution in the absence of calcium is also noted, i.e. the formation of a labile layer on the membrane surface, which is affected by the small cross-flow that exists inside hollow fibers, even in the nominally dead-end mode of operation.     

Preparation, characterization and effect of annealing on performance of cellulose acetate/sulfonated polysulfone and cellulose acetate/epoxy resin blend ultrafiltration membranes

Polymeric membranes based on cellulose acetate (CA)--sulfonated polysulfone blends at three different polymer compositions were prepared by solution blending and phase inversion technique, characterized and subjected to annealing at 70, 80 and 90 °C. The permeate water flux, separation of bovine serum albumin and its flux by the blend membranes before and after thermal treatment, have been compared and discussed. Similarly, CA and epoxy resin (diglycidyl ether of bisphenol-A) were blended in various compositions, in the presence and in the absence of polyethyleneglycol 600 as non-solvent additive, using N,N^′-dimethylformamide as solvent, and used for preparing ultraflltration membranes by phase inversion technique. The polymer blend composition, additive concentration, casting and gelation conditions were optimized. Blend membranes were characterized in terms of compaction, pure water flux, water content and membrane resistance. The effects of polymer blend composition and additive concentration on the above parameters were determined and the results are discussed.     

Determination of cake thickness and porosity during cross-flow ultrafiltration on a plane ceramic membrane surface using an electrochemical method

The thickness and the porosity of a deposit during ultrafiltration experiments are determined using an electrochemical method. Twenty microelectrodes are mounted flush to a ceramic plane membrane and maps of deposit thickness are determined for three inlet/outlet distributors configurations. Combining an electrochemical method and a step transient method, the determination of the thickness and the porosity of a particles deposit is performed. These local thickness and porosity values are analyzed thanks to wall shear stress local measurements obtained in a previous work [Sep. Sci. Technol. 37 (10) (2002) 2251]. The results emphasize the heterogeneity of the deposit thickness, especially in zones of low wall shear stress. Furthermore, the porosity values of the deposit are ranged between 0.3 and 0.8 as a function of the location at the membrane surface.     

Modification of polyethersulfone ultrafiltration membranes with phosphorylcholine copolymer can remarkably improve the antifouling and permeation properties

The synthesized phosphorylcholine copolymer composed of 2-methacryloyloxyethylphosphorylcholine (MPC) and n-butyl methacrylate (BMA), blended with polyethersulfone (PES), was used to fabricate antifouling ultrafiltration membranes. Water contact angle measurements confirmed that the hydrophilicity of the MPC-modified PES membranes was enhanced to certain extent. X-ray photoelectron spectroscopy (XPS) analysis verified the substantial enrichment of MPC at the surface of the MPC-modified PES membranes. The adsorption experiments indicated that the adsorption amounts of bovine serum albumin (BSA) on the MPC-modified PES membranes were dramatically decreased in comparison with the control PES membrane. Ultrafiltration experiments were carried out to investigate the effect of MPC modification on the antifouling and permeation properties of the PES membranes, it was found that the rejection ratio of BSA was decreased, the flux recovery ratio was remarkably increased, and the degree of irreversible fouling decreased from 0.46 to 0.09. In addition, the MPC-modified PES membranes could run several cycles without substantial flux loss.     

Simulation of an ultrafiltration process of bovine serum albumin in hollow-fiber membranes

This work presents a numerical simulation of an ultrafiltration process of bovine serum albumin in solution, using hollow-fiber membranes. Such membranes are constituted of tiny polymer cylinders disposed in a tube-and-shell arrangement. The concentrate flows through the interior of the fibers and the pure solvent is recovered in the shell, assuming perfect solute rejection. In modeling the process, the flow of concentrate inside the fibers was considered to be laminar, with constant density, viscosity and solute diffusivity. Axial diffusion and angular effects were ignored. The model combines the effect of concentration polarization and adsorption, which are the two main limiting phenomena in ultrafiltration processes. The pressure on the shell side was considered constant and inside the fibers a linear pressure profile, dependent on the axial position, was adopted. The solution of the problem was achieved with the method of orthogonal collocation, with adequate choice of the weight function in the radial direction. In the axial direction, a finite-difference method was used. The numerical results were compared with experimental data available in the literature.     

Synthesis and characterization of nanoporous polycaprolactone membranes via thermally- and nonsolvent-induced phase separations for biomedical device application

A family of crosslinked poly(ethylene glycol) diacrylate (XLPEGDA) materials was synthesized via free-radical photopolymerization of poly(ethylene glycol) diacrylate (PEGDA) solutions in water. These materials are potential fouling-resistant coatings for ultrafiltration (UF) membranes. PEGDA chain length (n = 10–45, where n is the average number of ethylene oxide units in the PEGDA molecule) and water content in the prepolymerization mixture (0–80 wt.%) were varied, resulting in XLPEGDA materials with water permeability values ranging from 0.5 to 150 L μm/(m² h bar). Generally, water permeability increased with increasing prepolymerization water content and with increasing PEGDA chain length. Moreover, water permeability exhibits a strong correlation with equilibrium water uptake. However, solute rejection, probed using poly(ethylene glycol)s of well defined molar mass, decreased with increasing prepolymerization water content and increasing PEGDA chain length. That is, there is a tradeoff between water permeability and separation properties. Finally, the fouling resistance of XLPEGDA materials was characterized via contact angle measurements and static protein adhesion experiments. From these results, XLPEGDA surfaces are more hydrophilic in samples prepared at higher prepolymerization water content or with longer PEGDA chains, and the more hydrophilic surfaces generally exhibit less BSA accumulation.     

Effect of surfactant agents and lipids on optical resolution of amino acid by ultrafiltration membranes containing bovine serum albumin

Ultrafiltration experiments for the optical resolution of racemic phenylalanine were performed in a solution system containing bovine serum albumin (BSA) and surfactant agents (Triton X-100, Tween 20, sodium dodecyl sulfate), lipid (phosphaticylcholine) and fatty acid (palmitic acid sodium salt). It was found that -phenylalanine preferentially existed in the permeate at pH 7.0 due to the binding of BSA to -phenylalanine in the feed and that the separation factors (=concentration ratio of -isomer to -isomer in the permeate) increased with a decrease in the BSA solution containing no additives and in the BSA solution containing Triton X-100 or Tween 20. The unusual tendency that the separation factors were less than unity was observed and the separation factors decreased with a decrease in the feed concentration of phenylalanine during the ultrafiltration containing the palmitic acid sodium salt or the phosphatidylcholine. This is caused by the fact that the binding constants of -phenylalanine to BSA are higher than those of -phenylalanine in the BSA solution containing the palmitic acid sodium salt or phosphatidylcholine. Since there were found conformational changes of BSA in the presence of palmitic acid sodium salt based on circular dichroism measurements of BSA solution, the conformational changes of BSA were attributed to the higher affinity of -phenylalanine to BSA than that of -phenylalanine in the BSA solution containing the palmitic acid sodium salt or phosphatidylcholine.     

Use of atomic force microscopy and fractal geometry to characterize the roughness of nano-, micro-, and ultrafiltration membranes

Membrane surface roughness alters the surface area accessible to foulants and may influence macroscopic properties, such as zeta potential. It is usually quantified by atomic force microscopy (AFM) at a single scan size. This would be appropriate if roughness is independent of scale. This study shows that the root-mean-square roughness, R_RMS, is scale (or scan size, L × L) dependent through the power law R_RMS = AL^3−D. The coefficient, A, is the roughness at a scan size of 1² μm². D is the fractal dimension that relates the increase in roughness to the increase in scan size. Values for A and D were determined for a range of micro- and ultrafiltration membranes using an AFM scan series covering at least three orders of magnitude in L. They were also determined for nanofiltration membranes by re-analysis of data in the literature. The results suggest that using the power law expression allows potentially greater discrimination among membrane types and provides a way to quantify membrane roughness over a range of scales. It was further observed that the coefficients A and D of PVDF membranes showed positive and negative correlations, respectively, with the molecular weight cut-off. Additionally, zeta potentials of PVDF membranes measured by the tangential streaming potential method became more negative with increasing A and more positive with increasing D, suggesting possible significant influence of roughness on hydrodynamic transport of ions.     

A new route for the fabrication of TiO2 ultrafiltration membranes with suspension derived from a wet chemical synthesis

Titania ultrafiltration membranes were successfully fabricated by a new route, which was directly derived from the nanoparticles suspension that was the intermediate product prior to dry and calcine in the synthesis of nanoparticle by a wet chemical method. The morphology and the crystal structure of the prepared membrane were analyzed by SEM and XRD. The effect of various dipping time on the membrane thickness was investigated. The rejection of the bovine serum albumin (BSA, 67,000 Da) was used to evaluate the separation characteristics of these membranes, and the relationship between the dipping time and the optimization thickness of the membrane was built on the base of the data of the pure water flux. SEM images showed that the surface of the membrane was defect-free and XRD revealed that the titania crystalline phase was pure anatase. The membrane thickness increased linearly with the square root of the dipping time and the dipping time of 30 s was necessary to form a defect-free titania layer on the top of supports. The titania layer derived from the dipping time of 30 s could be of thickness of 5.9 μm and an average pore size of 60 nm. The pure water permeability of the membrane was 860 × 10⁻⁵ L/(m² h Pa) (860 L/(m² h bar)), and the BSA rejections of the membranes prepared reached to 90% after 20 min running.     

The influence of protein aggregates on the fouling of microfiltration membranes during stirred cell filtration

Although macromolecular fouling of microfiltration membranes is one of the critical factors governing the performance of these filtration processes, there is still little fundamental understanding of the underlying phenomena that influence the initiation, rate, and extent of fouling. We have obtained experimental data for the flux decline during the stirred cell filtration of different commercial preparations of bovine serum albumin (BSA) through asymmetric polyethersulfone microfiltration membranes. The fouling characteristics of these commercial solutions varied substantially, with the flux decline directly related to the technique utilized to initially precipitate and prepare the BSA. Prefiltration of BSA solutions prior to microfiltration substantially reduced their fouling tendency, with the degree of improvement increasing as the prefiltration was performed through smaller molecular weight cut-off membranes. The protein solutions were also characterized using gel permeation chromatography (GPC), with the fouling tendency of the different BSA preparations highly correlated with the concentration of BSA dimers and other high molecular weight species present in these BSA solutions. These results suggest that BSA fouling of these microfiltration membranes is associated with the deposition of trace quantities of aggregated and/or denatured BSA, with these fouling species serving as initiation sites for the continued deposition of bulk protein.     

Effect of ionic strength and protein concentration on the transport of proteins through chitosan/polystyrene sulfonate multilayer membrane

The influence of ionic strength and protein concentration on the transport of bovine serum albumin (BSA), ovalbumin and lysozyme through chitosan (CHI)/polystyrenesulfonate (PSS) multilayers on polyether sulfone supports are investigated under ultrafiltration conditions. The percentage transmission and flux of BSA, ovalbumin and lysozyme were found to increase with increase in salt concentration in the protein. The percentage transmission of BSA through 9 bilayer membrane was found to increase from 5.3 to 115.6 when the salt concentration was varied from 0 to 1 M. It was observed that 0.1 M NaCl in BSA solution is capable of permeating all the BSA. When the salt concentration in BSA was further increased, a negative solute rejection (solute enrichment in permeate) was found to take place. With 9 bilayer membrane, the percentage transmission of ovalbumin was found to increase from 23.3 to 125.8 when the salt concentration in protein was increased from 0 to 0.05 M. The effect of protein concentration on protein transport is studied taking BSA as a model protein. BSA was rejected by the multilayer membrane at all the studied concentrations (0.25, 0.5, 1 and 2 mg/ml). With increase in feed concentration, maximum rejection of protein occurred at higher number of CHI/PSS bilayers. BSA solution flux was found to decrease with an increase in BSA concentration. This study indicates that it is possible to fine tune the transport properties of proteins through multilayer membranes by varying the concentration and ionic strength of protein solutions.     

Studies on cellulose acetate and sulfonated poly(ether ether ketone) blend ultrafiltration membranes

New ultrafiltration membranes based on chemically and thermally stable arylene main-chain polymers have been prepared by blending the sulfonated poly(ether ether ketone) with cellulose acetate in various compositions in N,N^′-dimethylformamide as solvent by phase inversion technique. Prepared membranes have been subjected to ultrafiltration characterizations such as compaction, pure water flux, water content, and membrane hydraulic resistance. The pore statistics and molecular weight cut-off (MWCO) of the membranes have been estimated using proteins such as trypsin, pepsin, egg albumin and bovine serum albumin. The pore size increased with increasing concentrations of sulfonated poly(ether ether ketone) in the casting solution. Similarly, the MWCOs of the membranes ranged from 20 to 69 kDa, depending on the various polymer compositions. Surface and cross-sectional morphologies of membranes were analyzed using scanning electron microscopy. The effects of polymer compositions on the above parameters were analyzed and the results are compared and discussed with those of pure cellulose acetate membranes.     

A constant flux based mathematical model for predicting permeate flux decline in constant pressure protein ultrafiltration

This paper discusses a novel approach for predicting permeate flux decline in constant pressure ultrafiltration of protein solutions. A constant pressure process is assumed to be made up of a large number of small, sequential, constant flux ultrafiltration steps: the flux decreasing due to fouling and other related factors at the end of each step. The advantage of this approach is that constant flux ultrafiltration is easier to study, characterize, and model than constant pressure ultrafiltration. Consequently model parameters can be obtained in reliable and reproducible manner. Constant pressure ultrafiltration is dynamic in nature since both the magnitude of osmotic back-pressure and the extent of membrane fouling decrease as the permeate flux decreases with time. The proposed model takes into consideration the interplay between permeate flux, concentration polarization, and membrane fouling. The model demonstrates that the initial rapid flux decline is due to a combination of concentration polarization and membrane fouling while during the remaining part of the process, the effect of concentration polarization becomes negligible. The model also shows that concentration polarization affects the initial flux decline only at higher transmembrane pressures. This model which was validated using experimental data is conceptually simpler than other available models and easy to use. In addition to its value as a predictive tool it would particularly be useful for deciding appropriate start-up conditions in ultrafiltration processes.     

The use of moment theory to interpret diffusion and sieving measurements in terms of the pore size distribution in heteroporous membranes

Moment theory has been applied to model porous membranes to show that one can place reasonable bounds on the cumulative pore size distribution, the hindered diffusivity or the reflection coefficient of large solutes in a heteroporous membrane by measuring the diffusive permeability to a small solute, the hydraulic permeability and one or two additional transport characteristics. These additional measurements involve either the flux of a small solute at Pe1, the hindered diffusivity of a large solute or the reflection coefficient of a large solute at Peå1. Membrane heteroporosity is incroporated in the predicted bounds without requiring one to make any a priori assumptions about the nature of the pore size distribution. In this paper, the results from calculations performed with different model membranes containing log-normal pore size distributions are reported. A comparison of the results obtained with three different membranes shows that one can distinguish between membranes with the same average pore size but different pore size distributions by measuring either the hindered diffusion coefficient or the reflection coefficient of two different sized solutes. A comparison of the bounds on D and the bounds on σ predicted from different types of transport measurements shows that, under certain conditions, one can place tighter bounds on one transport characteristic by measuring a different one.     

Fouling and cleaning of membranes in the ultrafiltration of the aqueous extract of soy flour

The aqueous extract of soy flour is an emulsion/suspension of proteins, lipids and carbohydrates. The foulant deposit formed on the surface of polysulfone membranes in the ultrafiltration of this complex extract was investigated from several aspects including thickness, physical structure, chemical analysis and rheological behaviour. SEM studies showed the thickness of the foulant deposit was approximately 0.2 μm for 50000 MWCO membrane and 0.4 μm for 100000 MWCO membrane. The structure of the foulant deposit consisted of lipids in a globular form of 0.2 to 1 μm diameter adhered to, and supported by, a protein-polysaccharide matrix. Rheological measurements were conducted on a sample of the foulant deposit collected from the 100000 MWCO membrane. This foulant deposit exhibited pseudoplastic and viscoelastic properties which totally resisted the surface shear stresses in the flat-plate module. Recovery of the water flux of the fouled membranes was achieved by a four-stage cleaning procedure comprising successive stages of washing with sodium hydroxide, protease detergent, sodium hypochlorite and flushing with water.     

Mechanisms and factors influencing the removal of microcystin-LR by ultrafiltration membranes

In this study, we investigate the application of ultrafiltration (UF) for the removal of the cyanotoxin, microcystin-LR, and determine the dominant removal mechanisms. System variables examined included membrane characteristics, feed concentration, water recovery and operating pressure. While adsorption dominated rejection for most UF membranes, at least at early filtration times, both size exclusion and adsorption were important in removing microcystin-LR by the tight thin-film (TF) membranes. Adsorption was primarily attributed to hydrophobic interactions, although hydrogen bonding and physical surface properties such as surface roughness, thickness, and porosity may also play a role. Polysulfone membranes, the most hydrophobic membrane examined, significantly adsorbed microcystin-LR (91%), whereas the more hydrophilic cellulose acetate membranes adsorbed little or no microcystin-LR. The initial feed concentration had a significant influence on the adsorption capacity of TF membranes for microcystin-LR, which could be described based on a linear adsorption isotherm. An increase in water recovery and/or operating pressure led to an increase in the adsorption of microcystin-LR, probably due to increased convective transport. On the other hand, microcystin-LR rejection through size exclusion was reduced for higher water recovery and/or applied pressure.