Search

Membrane fouling continues to hamper the performance of membrane‐filtration processes. A challenge with macromolecular foulants like proteins is that macroscopic characterizations, like net electrical charge, may be poorly correlated with membrane fouling. This necessitates a molecular‐scale analysis of the local interactions. In this study, molecular dynamics simulations have been performed to understand the interactions between two similar‐sized proteins with opposite overall charges (namely, lysozyme and α‐lactalbumin) and a negative‐charged membrane. Surprisingly, the protein–membrane distances and adsorption probabilities of both proteins are similar. Compared with the positive‐charged lysozyme, the negative‐charged α‐lactalbumin exhibits (a) greater protein–membrane attractive interaction energy due to synergy among adsorption sites; (b) lower root‐mean‐squared deviations (RMSD); and (c) greater number of residues that show low root‐mean‐squared fluctuations (RMSF). These results indicate that local interactions are critical and thus highlight the pitfall of using the overall protein characteristics as predictors of membrane fouling. [ABSTRACT FROM AUTHOR]

Molecular dynamics simulations were performed to understand the adsorption of positive‐charged lysozyme onto negative‐charged polyvinylidene fluoride membrane at three pH and two ionic strengths. The lysozyme was initialized at six orientations at 10 Å from the membrane and adsorption was deemed to occur when the lysozyme is less than 4 Å from the membrane. Local interactions are clearly important: (i) despite opposite net charges, no adsorption occurs in some cases throughout; (ii) lysozyme‐membrane separation distances differ among the initial orientations of lysozyme; and (iii) the correlation between lysozyme‐membrane interaction energy and adsorption probability is poor. Correspondingly, seven key local adsorption sites on lysozyme were identified. Of the two most common sites, one anchors due to electrostatics, while the other is not electrostatics‐based. Also, different sites are dominant in different feed conditions and have different interaction energies. These results reveal the importance of local interactions in membrane fouling by proteins in different feed conditions. [ABSTRACT FROM AUTHOR]

Graphical abstract Highlights • Evolution of oil-emulsion cake layer on HF membranes observed in real time. • At the same permeate flux, TMP of oil emulsion was higher than latex particle. • Oil gave a thinner cake layer that was denser and moving with the cross flow. • Drastic fluctuations of cake thickness and porosity versus permeate volume for oil. • Greater oil-oil and oil-membrane interaction energies underlie the fouling observed. Abstract In view of membrane-filtration being a promising means for treating the large volumes of oily wastewater, much efforts have been dedicated to understanding membrane fouling by oil emulsions. However, the understanding on cake fouling, which is the governing form of fouling throughout filtration, by oil emulsions remain poorly understood. To bridge this gap, this study focused on characterizing the evolution of the oil-emulsion cake layer characteristics on hollow fiber (HF) membranes in real time using the direct observation through the membrane (DOTM) technique. The more well-studied latex particles were studied alongside to provide a benchmark for comparison. Despite the same feed concentration and crossflow velocity used, the transmembrane pressure (TMP) and cake layer resistance magnitudes were greater for oil emulsion than for latex particles at the same permeate flux. DOTM images revealed that this was linked to the following behaviors: (i) oil tended to deposit as a moving cake layer, whereas latex as scattered deposits; and (ii) oil gave a thinner yet denser cake layer. Moreover, it was observed that the cake thickness and porosity fluctuated significantly with respect to permeate volume for the oil emulsions, because the moving cake layer gave greater variations of cake thicknesses along the membrane. Finally, the interfacial interaction energy predicted by the XDLVO model gave higher oil-membrane and oil-oil affinities compared to that of latex, which underlie respectively the more extensive fouling and denser cake exhibited by oil. The unique characteristics of the cake layer of oil emulsion vis-à-vis the well-studied particulate foulant (i.e., latex) shown here provides new perspectives on the monitoring and control of oil fouling. [ABSTRACT FROM AUTHOR]

A novel inverse fluidized bed (IFB) was developed to improve membrane fouling mitigation vis-à-vis the conventional aeration method. The fluidized media, whose density was less than water and oleophilic, were more effective than bubbles for both feeds containing oil and particulates due to their greater inertia. The key highlights are (i) for all the conditions investigated, IFB consistently gave a lower trans-membrane pressure (TMP) than aeration at the same energy requirement; (ii) IFB performed better even when the energy input was reduced by 87.5%; (iii) IFB provided an impressive enhancement of up to 54 times that of dead-end filtration. [ABSTRACT FROM PUBLISHER]

Membrane distillation (MD) can concentrate non-volatile solutes or remove volatiles and dissolved gases from an aqueous feed. A microporous hydrophobic membrane provides a barrier between the hot feed and cold distillate. Although MD can operate at ambient pressure and moderate temperatures, use waste heat, and treat wastewater via an MD-bioreactor, it has problems such as temperature polarization, liquid weeping to the distillate side, and membrane fouling. Prior studies speculated that fouling can add a heat- or mass-transfer resistance, or cause a vapor-pressure reduction owing to the Kelvin effect, but did not isolate these effects. This study confirms that the vapor-pressure depression owing to the concave interface in the small pores of the fouling layer is a dominant cause of the 25–63% flux reduction observed for humic-acid fouling on PTFE and PVDF membranes. This study underscores the importance of selecting MD membranes based on their pore-size distribution rather than just their nominal diameter in order to maximize the contribution of Knudsen diffusion. It suggests the development of dual-layer membranes having a thin hydrophilic layer with relatively large pores overlying a hydrophobic layer with a typical MD membrane structure in order to mitigate the vapor-pressure reduction owing to membrane fouling. [ABSTRACT FROM AUTHOR]

Membrane fouling remains a tremendous obstacle in the implementation of membrane technology, which has motivated the use of a wide range of experimental techniques to study and understand fouling behavior in the past decades. However, molecular-level insights, which underlie the macroscopic phenomena observable experimentally-, remain incomplete. With the rapid advancement of computational power, molecular dynamics (MD) simulation has become increasingly popular to explore the interaction occurring at the ternary interface (i.e., foulant, membrane and solvent) by using explicit atom representations. MD studies have quantified interfacial physical properties and generated MD trajectories for direct visualization of the adsorption process. This review focuses on reported MD studies on membrane fouling and related foulantadsorption processes. The fundamentals of MD and procedures for constructing an MD system for fouling investigation are overviewed first. Subsequently, representative parameters and related insights generated from unbiased simulation are discussed. Then, biased simulations with enhanced sampling algorithms, which addresses some shortcomings of unbiased simulations and allows for quantification of the free energy landscape, are outlined. Additionally, the unique characteristics of representative foulants revealed by MD are summarized. Finally, future perspectives are proposed to enhance the value of MD as a tool for understanding and predicting membrane fouling. [Display omitted] • Review the use of molecular dynamics (MD) simulations in membrane fouling studies. • Overview fundamentals of MD and procedures for constructing MD systems. • Discuss representative parameters and insights generated from unbiased simulations. • Outline biased simulations with enhanced sampling to obtain free energy landscape. • Summarize unique characteristics of representative foulants revealed by MD. [ABSTRACT FROM AUTHOR]

Abstract: Membrane distillation (MD) is a continuous process whereby a hot aqueous feed provides the latent heat to vaporize water that diffuses through a hydrophobic microporous membrane and condenses. MD is attractive for saline and wastewater treatment as well as concentrating heat-sensitive solutions such as fruit juices and biological fluids since it operates at moderate temperatures and atmospheric pressure, rejects particulates and nonvolatile solutes, uses non-selective membranes, and has low capital costs. However, MD and MD bioreactor (MDBR) performance is compromised by fouling. The effect of fouling layers with reasonably large pores (>50nm) is reasonably well-understood and has been incorporated into MD models. However, the effect of fouling layers with very small pores or free volume (<50nm) owing to macromolecular- and bio-fouling (MMBF) is not well-understood. While MMBF introduces an additional resistance to heat transfer, it can also reduce the vapor-pressure driving force. This latter effect is unique to MD fouled by MMBF. In this analysis the vapor pressure reduction has been incorporated into a model that indicates the large flux declines observed in prior studies can be explained by the vapor-pressure reduction associated with pore diameters ranging from 3.9 to 8.5nm, which agree well with evapoporometry characterization of MMBF sludges. The predicted flux decline and temperature polarization coefficient indicate that the vapor-pressure reduction increases markedly for effective pore diameters less than 10nm, but can be mitigated by increasing the thermal resistance of the membrane and by fabricating a dual-layer membrane for which the feed side is hydrophilic with relatively large pores. [Copyright &y& Elsevier]

More than two decades after the highly-cited review on microfiltration of suspensions and macromolecular solutions, it is time to revisit and update the fundamentals and applications of this topic. This includes addressing inevitable fouling phenomena, the description of fouling, fouling models, means to mitigate fouling and the modernization of membrane modules. Mass transfer limitations, related to concentration polarization and fouling, remain the Achilles' heel. Since empirical expressions that describe fouling offer limited predictive ability and mechanistic insight, there is an urgent need for more in-depth fundamental understanding. Elucidating interfacial interaction energies has gained much attention in recent years; the DLVO model (and its extension) has been validated for a wide range of foulants. The use of molecular dynamics simulations to model transport and fouling represents a major advance, providing insights not possible through experiments. Building on classical fouling mechanisms, new models hold promise to provide more realistic description of fouling phenomena. Fouling mitigation efforts have focused on high-throughput modification platforms and more energy-efficient unsteady-state shear methods. Finally, membrane modules can be modernized based on the knowledge accumulated to better improve their efficiency, and factoring in important considerations like feed type, mechanical stability, hydrodynamics, economics and application goals. Image 1 • Describes advances in microporous membranes, fouling, analytical model and module. • Reveals that high throughput and machine learning will expedite optimization. • Indicates that interaction energy analysis is critical for better fouling prediction. • Demonstrates combined fouling models can often predict fouling accurately. • Reports that unsteady flow reduces concentration polarization/fouling. [ABSTRACT FROM AUTHOR]

Abstract The evolution of the cake during dead-end microfiltration of different compositions of bidisperse feeds was investigated using network modeling and three-dimensional (3D) optical coherence tomography (OCT) imaging. The foulant particles were latex of two different sizes (namely, 3 and 5 µm) and the membrane was polycarbonate track-etched with nominal diameter of 2 µm. The results from the network model indicate that (i) the deposition of the smaller particles on the non-porous area was reduced by the presence of larger particles, whereas that of the larger particles was not affected by the concentration of the smaller ones; and (ii) monodisperse small particles gave a greater specific cake resistance than monodisperse large particles, and the specific cake resistance of the bidisperse mixture can only be reduced with a sufficient proportion of larger particles. OCT results substantiated the model by revealing that (i) bidisperse feeds tended to form homogeneous cakes, but the presence of a higher concentration of large particles led to heterogeneous cakes; (ii) the presence of a lower concentration of larger particles enhanced clustering, while a higher concentration reduced that; and (iii) in the longer-term filtration, the monodisperse feed of small particles dynamically deposited and detach continuously, while the bidisperse feed with the highest concentration of larger particles exhibited negligible deposition and detachment. Highlights • Studied fouling mechanisms and cake formation by various bidisperse mixtures. • Network model revealed different deposition on nonporous area and cake resistance. • OCT revealed different cake growth, clustering and dynamic behavior. • Cake resistance only lowered above a critical proportion of large particles. • Bidisperse feeds tended to form homogeneous cakes. [ABSTRACT FROM AUTHOR]

The anaerobic fluidized-bed membrane bioreactor (AnFMBR) makes use of fluidized granular activated carbon (GAC) to mechanically scour the membrane to mitigate fouling. The question addressed in the current study is whether fluidizing bidisperse mixtures (i.e., particles of two different sizes) can improve the energy-efficiency in mitigating membrane fouling. The hydrodynamic parameters (including particle velocity, particle concentration, particle momentum and extent of segregation) of the fluidized media numerically obtained via a Two-Fluid Model (TFM) were correlated with the fouling rate experimentally obtained. Three bidisperse mixtures with the same mean particle diameter (namely, 1 mm) but different particle diameter ratios (namely, d p2 /d p1 = 1.22 – 1.86) were studied over a range of superficial liquid velocities. The key highlights are that (i) the fouling rate decreased (i.e., fouling mitigation was more effective) as the particle diameter ratio increased, which indicates the benefits of fluidizing multi-sized particle mixtures over monodisperse ones, (ii) the fouling rate (dTMP/d t ) was positively correlated to the mixing index ( M ), which implies that segregated systems were more effective at mitigating membrane fouling, and (iii) the effect of bidispersity was more significant at the lower power input, which suggests that the advantages of fluidizing such bidisperse mixtures should only be harnessed at moderate power but not higher ones. [ABSTRACT FROM AUTHOR]

Fouling control by bubbling is widely implemented in membrane bioreactors (MBRs), but a mechanistic understanding of the relationship between the bubble characteristics and critical flux is missing. This study is targeted at bridging the gap. A microfiltration setup employing bubbling as a means of fouling mitigation was set up. The direct observation through the membrane (DOTM) technique was used to determine the critical flux of a feed containing micron-sized polystyrene particles, while a high-speed video camera was used to characterize the bubble behavior. The results indicate that (i) the local critical flux ( J c,local ) values increased with height along the membrane and with gas flow rate, but was relatively invariant with liquid flow rate; (ii) with respect to height, the mean bubble velocity and mean bubble area per bubble increased, while the number of bubbles per unit membrane area decreased; (iii) bubble momentum and mean area per bubble had the most positive correlation with J c,local ,; and (iv) the overall critical flux ( J c,overall, which is the average of the local critical fluxes) increased linearly with respect to the ratio of gas to total flow rates, and steeply then more gradually with respect to power required. Finally, a comparison between the two two-phase flow systems employed for fouling mitigation (namely, bubbling and fluidized granular activated carbon (GAC)) was carried out. [ABSTRACT FROM AUTHOR]

The anaerobic fluidized bed membrane bioreactor (AnFMBR) utilizes the fluidization of granular activated carbon (GAC) particles as an unsteady-state shear means for mitigating membrane fouling. In this study, a two-fluid model (TFM) was validated against experimental results and used to obtain a comprehensive landscape of local water and particle velocities and concentrations throughout the reactor. The results provided interesting insights. Firstly, both water and particle velocities tended to be non-uniform vertically and laterally, which indicate that the extent of fouling across the membrane would be non-uniform. Secondly, to attain the same water shear stress or particle momentum values, the power required was greater for the larger particles, which indicate that smaller particles were more energy-efficient. Thirdly, at the same power input, the water shear stress or particle momentum values for the different particle diameters overlapped, which indicates that other parameters can be manipulated to improve performance even when particle diameter has been fixed. Fourthly, generally negative correlations existed between fouling rate and the hydrodynamics of both phases, but fouling rate and power input were poorly correlated, which suggests surprisingly that increasing power would not improve the fouling mitigation. [ABSTRACT FROM AUTHOR]

In view of the proliferating applications involving membrane-based separation, the inevitable membrane fouling phenomenon is well-studied. However, the influence of foulant shape on membrane fouling remained incompletely understood. Therefore, the goal of this study was to investigate the fouling behaviors by three different shapes of latex particles (namely, peanut, pear and sphere) with similar mean diameters in the range of 3.5–3.8 μm. External and internal fouling were studied using polycarbonate track-etched (PCTE) membranes with nominal pore sizes of 2 and 8 μm, respectively. Flux data were taken, and fouling models, optical coherence tomography (OCT) and field-emission scanning electron microscope were employed. Regarding external membrane fouling, the sphere-shaped particles displayed the worst flux decline, which is tied to the greatest tendency to deposit on the membrane pores, as revealed by the network fouling model. As for internal fouling, the flux decline curves were more similar among the particle types. The fouling model shows the sphere-shaped particles gave the densest internal cake, which caused slightly steeper flux decline in the latter part of the filtration. Furthermore, OCT indicates that the highest fouling fractions at 7.5 μm below the membrane were exhibited for the peanut-shaped particles, which is due to the preferential vertical orientation resulting in deeper penetration. Results here highlight the non-negligible influence of foulant particle shape, which warrants further understanding and accounting for membrane fouling studies. [Display omitted] • Used OCT to reveal influence of foulant particle shape on membrane fouling. • Spheres were more prone to pore-blockage and thus gave worst flux decline. • Non-spherical particles gave less dense internal cakes due to rotational effects. • Peanut-shaped particles penetrated deeper due to preferential vertical orientation. • Particle shape affected flux decline more when external fouling was dominant. [ABSTRACT FROM AUTHOR]

Membrane fouling persists to be among key obstacles in liquid membrane-based separation processes, and thereby the underlying understanding has actively been pursued. The two gaps attempted to be addressed here are namely, (i) internal fouling, which has limited studies due to limitations of instruments available; and (ii) fouling by mixtures of oppositely charged particles, as an extension to the studies on effect of surface charge of pure foulants on fouling. With the sub-micron pure constituents (namely, negatively charged carboxylated latex particles and positively charged aminated latex particles) as benchmarks, mixtures of three different compositions of the two particle types were investigated. The membrane was polycarbonate track-etched (PCTE) membrane. Surprisingly, the mixtures gave worse flux decline than that by purely positively charged aminated latex, which contradicts electrostatic considerations. The fouling model indicates that the mixtures exhibited higher pore blockage parameters (α) than either of the pure constituents, which agree with the flux decline trends. On the other hand, the mixtures gave internal cake resistance parameters (R c /R m) in between that of the pure constituents, which disagree with the fouling trends. Furthermore, optical coherence tomography (OCT) analysis indicates that the mixtures had taller buildups on the membrane as well as heterogenous internal cakes. Therefore, the steepest flux declines for the mixtures seem to be better correlated with the extensive external fouling as manifested in the taller buildups on the membrane pores, rather than internal fouling since the internal cakes were more porous. The contradictory extensive external and mitigated internal fouling phenomena are unexpected based on electrostatic considerations, and thereby highlight that fouling resulting from mixtures may not be readily extrapolated from understanding based on pure particle feeds. Image 1 • Steeper flux decline by mixtures of oppositely charged constituents than pure ones. • Mixtures gave greater pore blockage than constituents, agreeing with flux trends. • Mixtures gave intermediate internal-cake resistance, deviating from flux trends. • OCT trends affirm taller external cake and porous internal cake by mixtures. • Contradictory extensive external fouling and mitigated internal fouling by mixtures. [ABSTRACT FROM AUTHOR]

• Fluidizing GAC and fiber looseness reduced both reversible and irreversible fouling. • Fiber looseness enhanced momentum transfer from GAC particles to membrane. • Larger molecular weight organics more effectively prevented from depositing. • GAC scouring effective in reducing fouling by microbial by-product-like substances. • GAC fluidization reduced fouling layer roughness, which further mitigated fouling. Membrane fouling mitigation by fluidized granular activated carbon (GAC) is well-acknowledged to be effective in mitigating external reversible fouling. However, the effect of fluidized GAC and fiber looseness on the alleviation of irreversible fouling is not known, which motivated the current study. Analyses included reversible and irreversible membrane fouling resistance, GAC particle velocities, foulant characteristics, permeate quality and membrane morphology. Results indicate that (i) membrane fouling resistances were significantly reduced by GAC fluidization and fiber looseness, with reversible and irreversible fouling respectively reduced by up to 17 and 35 times; (ii) advantages conferred by fiber looseness was tied to the decreased GAC particle velocities by about 10%, which implies a greater transfer of momentum from the fluidized GAC to the membranes; (iii) GAC scouring reduced 87% of reversible fouling and 48% of irreversible fouling, largely by microbial by-product-like substances but not so much by humic acid-like substances; whereas fiber looseness had a greater impact on the composition of the irreversible foulants; (iv) fluidized GAC particles and fiber looseness were more effective in alleviating the deposition of higher molecular weight organics in reversible fouling; (v) not only was the membrane integrity not compromised by the scouring GAC, but better permeate quality (i.e., 40% lower DOC in the permeate) was achieved; and (vi) roughness of the fouled membrane surface was 63% lower in the presence of GAC fluidization, which further reduced membrane fouling. This study thereby provides valuable insights towards the better design and operation of such fouling mitigation means for wastewater treatment. [ABSTRACT FROM AUTHOR]

In view of the inevitability of membrane fouling, it is critically important to understand the underlying mechanisms to alleviate fouling and thereby improve the feasibility of membrane-filtration for more applications. This study was targeted at understanding the internal membrane fouling mechanisms by three types of polystyrene (PS) particles (sized at ~ 0.45 μm; different surface charges) during microfiltration with the same polycarbonate track-etched (PCTE) membrane (nominal pore size of 2 μm). Optical coherence tomography (OCT) technique, which was employed to monitor membrane fouling, indicated greater extents of particle deposition in the case of the positive and less negative PS particles, which is consistent with the more severe flux declines relative to the more negative PS particles. The new fouling model developed that additionally accounts for internal cake filtration, on top of the well-reported pore constriction and pore blockage, revealed the fouling mechanisms at play. For the positive PS particles, the attractive particle-membrane interaction led to a more homogeneous layer-by-layer cake growth, which resulted in denser packing in the pore and higher internal cake resistance, and thereby the worst flux decline. In contrast, for the negative PS particles, the repulsive particle-particle and particle-membrane interactions resulted in more random deposition of the particles inside the pores, resulting in more extensive pore constriction and blockage, as well as lower resistance of the internal cake. This study underscored the importance of surface charge in internal fouling and provided insights on the lesser reported internal fouling mechanisms. • Internal fouling of PCTE membrane by PS particles of different charges studied. • Flux decline trends affected by zeta potentials of foulants despite the same sizes. • OCT found more internal deposition by positive and less negative PS particles. • New internal fouling model showed importance of internal cake filtration mechanism. • Greater flux decline linked to higher internal cake resistance. [ABSTRACT FROM AUTHOR]

Membrane fouling in organic solvent environments remains poorly explored despite its significance in chemical and pharmaceutical industries. This study uses molecular dynamics (MD) simulations and experiments to explore lysozyme fouling in water, as well as four organic solvent environments, namely, 30 % v/v and 50 % v/v isopropyl alcohol (IPA), and 30 % v/v and 50 % v/v dimethyl sulfoxide (DMSO). Experimentally, flux declines were least with IPA and worst with DMSO. Biased simulations indicate the worst fouling in DMSO is tied to the most attractive lysozyme-membrane energy in the presence of DMSO. However, the relative attractive energies for IPA and water do not agree with the relative flux declines, indicating other factors are more influential when the interaction energies are similar. To understand the gentler flux decline for IPA despite the more attractive lysozyme-membrane energy, radial distribution functions (RDFs) were obtained from unbiased simulations. Analyss of the water and solvent films around both the membrane and the lysozyme molecule reveal that the denser water film around both entities induced by the presence of IPA serves as a barrier for fouling and thus leads to less flux decline. The results underscore the complexity of fouling in organic solvent systems, cautioning against direct use of the understanding based on aqueous systems. [Display omitted] • Membrane fouling by lysozyme in water, IPA and DMSO studied. • Flux decline worst in DMSO and least in IPA. • Worst fouling in DMSO tied to most significant lysozyme-membrane attraction. • Least fouling in IPA tied to denser water film around membrane and lysozyme. • Both biased and unbiased molecular dynamics simulations needed for full explanation. [ABSTRACT FROM AUTHOR]

Graphical abstract Highlights • Spacer vibration could significantly alleviate membrane fouling. • 3-D spacer reduced more fouling compared to 1-D and 2-D spacers. • Increasing vibration frequency and amplitude enhanced fouling control. • Intermittent vibration can be adopted for effective fouling control. • Spacer vibration consumes less energy compared to gas bubbling. Abstract In this study, a novel turbulence promoter based on vibrating spacers was investigated to control fouling in a submerged flat sheet membrane filtration system. The impacts of spacer types (1-D, 2-D, and 3-D), parallel distance of spacers to the membrane, vibration frequency and amplitude, and intermittent vibration interval on membrane fouling potential were examined. The results showed that vibration of 3-D spacers (sinusoidal geometry) could achieve up to 48% and 25% more fouling reduction than 1-D (flat plate) and 2-D (flat plate with grooves) spacers respectively when the spacers were located at a distance of 0.1 mm to the membrane. Increasing vibration frequency of the 3-D spacer from 1 to 2.5 Hz and increasing amplitude from 0.8 to 2 cm improved membrane fouling control efficiency by 33% and 34% respectively. The improved membrane performance was attributed to combined effects of the enhanced turbulence kinetic energy associated with back-transport of foulants and shear rate on the membrane surface as illustrated by the three-dimensional computational fluid dynamics (3-D CFD) simulations. An experimental and power consumption comparison of spacer vibration and gas sparging for fouling control was also performed. This study highlights the feasibility of vibration spacers in enhancing submerged flat sheet membrane performance. [ABSTRACT FROM AUTHOR]

Highlights • Studied the striping phenomenon whereby the foulants deposited as regular streaks. • The direct-observation DOTM technique employed to characterize the stripes. • Stripes formed by oil drops stabilized by Tween surfactants but not other surfactants. • Stripes easier to remove from membrane than the more uniform layer of oil droplets. • Among the three Tween surfactants, the striping characteristics were largely similar. Abstract The striping phenomenon in membrane filtration, whereby the foulants deposit as regular streaks rather than a more uniform layer, was first observed more than 30 years ago (Jonsson, 1987), but the understanding has remained limited to a few subsequent studies (Henriksen and Hassager, 1993; Larsen, 1991; Li et al., 2016; Tanudjaja et al., 2017; Tarabara et al., 2002). In view of the potential practical implications of the stripes in terms of membrane fouling and fouling mitigation, this study was targeted at an in-depth characterization of the stripes. The direct observation through the membrane (DOTM) technique was employed to observe the conditions when the stripes formed by oil emulsions stabilized by three Tween surfactants at a lower cross-flow velocity, and higher oil concentrations and permeate fluxes. The results indicate that (i) other than hydrodynamic factors (e.g., permeate drag, tangential shear), foulant-membrane and foulant-foulant interactions played a role in stripe formation or disappearance, as evident in the formation of stripes only by the oil emulsion stabilized by the Tween surfactants and the effect of pH; (ii) stripes made up of oil droplets appeared to be easier to remove than the more uniform layer of oil droplets, as evident in the shorter time taken for the former to detach; and (iii) among the three Tween surfactants, the striping characteristics investigated were largely similar, except for the time taken for the deposits to detach. These are expected to have implications for fouling control and mitigation. [ABSTRACT FROM AUTHOR]

Abstract Membrane-based filtration is promising for the treatment of oily wastewater with stable micron-sized oil droplets, but is unfortunately limited by the inevitable membrane fouling phenomenon. To advance the understanding on membrane fouling by oil emulsion, this study made use of the direct observation through the membrane (DOTM) technique to experimentally visualize the evolution of fouling and determine critical flux, and also molecular dynamics simulations to unveil the interfacial interactions and behaviors underlying the different fouling behaviors. Three oil emulsion types with similar mean droplet sizes were studied, namely, one without surfactant, one stabilized by sodium dodecyl sulfate (SDS; a negatively charged surfactant) and one stabilized by dodecyltrimethylammonium bromide (DTAB; a positively charged surfactant). DOTM results indicate that the critical flux was the highest in the absence of surfactant and lowest for the DTAB-stabilized ones, while simulation results indicate that the interaction energies are clearly different among the different oil emulsion types. Both hence affirm that the presence of surfactant and different surfactant type distinctly changes the fouling tendencies. The key conclusions on the different fouling tendencies among the three oil emulsion types are summarized as follows. Firstly, the highest critical flux in the absence of any surfactant is linked to greatest oil-water interaction and least oil-membrane interaction, both of which cause the oil molecule to prefer being in the bulk aqueous feed. The radial distribution function (RDF) further substantiates this. Secondly, the higher critical flux exhibited by the SDS-stabilized oil emulsion compared to the DTAB-stabilized ones could be attributed to the negative charge of the former. Simulations affirm the comparatively greater oil-membrane repulsion through the RDF profile, lower oil-membrane interaction and higher oil-water interaction of the SDS-stabilized oil emulsion. Thirdly, despite surfactants generally have a stabilizing effect on oil emulsions, DOTM images show that coalescence was most extensive for the DTAB-stabilized oil emulsion, because of the greatest oil-membrane attraction and least oil-oil repulsion. Coupling both experiments and simulations, this study enhanced the mechanistic understanding on the effect of surfactant on membrane fouling by emulsions of the same oil. Highlights • Three oil emulsion types in the absence and presence of surfactant studied. • DOTM used to quantify critical flux and visualize evolution of fouling. • Presence of either surfactant reduced the critical flux of the oil emulsion. • Molecular dynamics simulation revealed different interaction energies. • Coalescence worst for DTAB-stabilized emulsion due to the least oil-oil repulsion. [ABSTRACT FROM AUTHOR]

The use of fluidized granular activated carbon (GAC) in anaerobic membrane bioreactors (AnMBR) as an unsteady-state shear means for mitigating membrane fouling is acknowledged to be effective and energy-efficient. Reports on both the lab-scale and pilot-scale have emanated, but the hydrodynamics of the fluidized GAC, which is unsurprisingly found to be closely tied to the extent of fouling mitigation, has only been studied on the lab-scale. The current study therefore was aimed at evaluating the feasibility of using a non-invasive accelerometer for a practical pilot-scale AnFMBR, in order to provide a means for real-time monitoring of such processes. A lab-scale AnFMBR was first investigated, and the efficacy of the accelerometer to differentiate between various positions along the reactor (i.e., different heights and sides) and operating conditions (i.e., superficial liquid velocity and overall GAC concentration) was proven. Then, the same effects on an idealized pilot-scale AnFMBR with new membranes and DI water was assessed, which also indicates that the accelerometer signals were able to reveal the hydrodynamics of the fluidized GAC in the reactor. Finally, in a practical pilot-scale AnFMBR with wastewater the accelerometer was shown to be able to diagnose the de-fluidization of the GAC, due to severe clogging issues, which thereby eliminated the fouling mitigation capacity. Trends in the trans-membrane pressure (TMP) and flux data were deficient in indicating the de-fluidization of the GAC. [ABSTRACT FROM AUTHOR]

Membrane-based filtration is an emerging technique for processing oily wastewater. Improving the efficiency of oil-water separation via membranes entails novel techniques for studying the complex interactions between the oil droplets and membrane underlying fouling. This paper presents the first study that applies optical coherence tomography (OCT) to the characterization of membrane fouling by oil emulsions. A series of dead-end filtration experiments was performed to characterize the rejection of oil droplets (~ 10 µm; hexadecane) by the membrane structure (0.45 µm PVDF) via three-dimensional (3D) OCT scanning in real time. The experimental results were compared with the control experiments with ~ 10 µm glass beads to identify the optical artifacts. Both the external and internal fouling by the oil droplets were successfully revealed by analyzing the variation in OCT intensity at various layers that were mathematically defined in terms of the coordinate surfaces parallel to, and above and below, the feed-membrane interface. The evolution of membrane fouling was quantified by evaluating the fraction of fouling voxels as a function of time at varied depths. This study demonstrates that the OCT-based characterization has the potential to shed light on the complex interactions occurring in oil-water separations via membrane filtration, particularly by providing real-time non-invasive monitoring of internal fouling, the understanding of which is valuable for both fundamental research and practical applications. [ABSTRACT FROM AUTHOR]

Microfiltration is an attractive means for treating oily wastewater, especially when the size of the oil droplets are micrometer-sized since the conventional techniques become deficient. A systematic study on the critical flux of oil-in-water emulsion, which behaves differently from other colloidal foulants with regards to deformation, coalescence and splitting, has not been carried out to date. This was the goal of the current study, which employed the Direct Observation Through the Membrane (DOTM) technique to characterize the critical flux of oil-in-water emulsions of various concentrations, and at various cross-flow velocities (CFV) and salt concentrations. Five observations can be highlighted here. Firstly, the oil droplets with a mean droplet diameter of approximately 5 µm exhibited critical fluxes equal to or greater than latex particles of 10 µm. This is likely due to the twin effects of membrane oleophobicity promoting back-transport of the oil foulant from the membrane and the presence of a droplet size distribution with larger drops that can enhance the shear-induced diffusion of the average drops. Secondly, the critical flux values did not agree with the model that is valid for the size range the mean droplet diameter falls in, but instead agreed with the model adapted for smaller particulate foulants. Thirdly, the increase in the critical flux with CFV was more significant for the lower oil concentration. Fourthly, a striping phenomenon was observed at higher oil concentrations and lower CFV values. Striping was not observed for latex particles. Fifthly, the critical flux decreased with salt concentration. These findings highlight the unique fouling behavior of oil-in-water emulsions in microfiltration. [ABSTRACT FROM AUTHOR]

The efficiency of membrane-based separations is limited by various fouling phenomena, which necessitates a mechanistic understanding in order to improve such processes. This study proposes a network-based approach, in which the membrane is discretized and each particle is individually monitored, to explore the underlying fouling mechanisms leading to pore blockage and cake growth during a membrane filtration process. In particular, the network-based approach provides more spatial resolution than the continuum approach, thereby gives a more in-depth insight into membrane fouling. The network model developed involves the construction of a two-dimensional (2D) network of pores to represent the membrane, a series of probabilistic criteria to describe the fate of each individual particle, and finally a protocol for evaluating the fouling during either a constant flux or constant TMP (transmembrane pressure) filtration. Three fouling parameters can be obtained to characterize the fouling behavior, namely, the probabilistic factor for deposition in the dead zone ( β ), the initial cake resistance ( R c 0 ), and the specific cake resistance with respect to cake thickness ( R ′ c ), by best-fitting the experimental flux-decline data to the network model. The capability of the network model to account for the topological and stochastic aspects of a fouling process provides for a more mechanistic understanding of the complex interactions between the fluid flow, membrane, and foulant particles vis-à-vis a model based on the continuum assumption. Although limited fouling cases were examined in the current study, it is expected that the network model developed here can be readily applied to study more complex phenomena involved in a membrane filtration process (e.g., shear-induced diffusion), and the associated insights would be significantly enhanced when coupled with more advanced fouling characterization techniques (e.g., Optical Coherence Tomography). [ABSTRACT FROM AUTHOR]

Membrane fouling by particulates is largely driven by permeate drag that causes the foulants to move towards the membrane. Cross-flow is used to induce shear at the membrane surface to mitigate the fouling. This study describes the flow-field mitigation of membrane fouling (FMMF) technique that reduces the fouling based on a judicious manipulation of the flow-field by inclining the channel walls to counter the permeate drag experienced by the foulants. The key mechanism is the additional transverse (i.e., orthogonal to the fluid flow direction) fluid vector caused by a small inclination of the channel walls (on the order of one to a few degrees). Both simulation and experiments confirm the efficacy of FMMF in mitigating fouling at a reduced energy requirement compared with the conventional channels with parallel walls. Simulations over a range of permeate fluxes indicated that a slight inclination angle of 1.15° can give a deposition factor less than that in the conventional channel with parallel walls even at twice the permeate flux. Direct observation through the membrane (DOTM) experiments for fouling using two particle diameters over a range of power inputs showed that the critical fluxes were significantly increased in the FMMF module compared with the conventional module. This study provides a proof-of-concept of the principle underlying FMMF and underscores the potential benefits. The direction for further development and scale-up is suggested. [ABSTRACT FROM AUTHOR]

The mitigation of membrane fouling due to the mechanical scouring behavior of granular activated carbon (GAC) particles of diameters ranging from 0.9 to 4.6 mm was studied using a GAC tangential shear-membrane filtration cell, wherein GAC was directed to scour the membrane surface by a magnetic stirrer placed 10 mm above the membrane. Two membrane pore sizes (namely, 0.1 and 0.45 μm), two permeate fluxes (namely, an above- and below-threshold flux) and two operation modes (namely, continuous and intermittent filtration) were investigated. Results indicated that: (1) GAC marginally improved threshold flux values, with the benefits more significant for the larger-pore membrane; (2) For continuous filtration, the presence of GAC reduced the cake layer resistance, but increased the irreversible resistance; (3) For intermittent filtration, GAC was surprisingly less effective in reducing cake layer resistance compared to continuous filtration for the smaller-pore membrane, but did not increase the irreversible fouling resistance; (4) Both threshold flux and fouling layer resistance were independent of GAC particle diameter, which contradicts the inertial lift model and thereby affirms the dominance of the mechanical scouring behavior studied here; (5) A model developed, based on momentum transfer between the GAC particle and foulant, provided a mechanistic understanding. [ABSTRACT FROM AUTHOR]

Backwashing is a common method for fouling mitigation. However, its impact on the pore-size distribution (PSD) of hollow fiber (HF) membranes has not been studied to date. This study quantitatively assessed the effects of filtration and backwashing cycles on the PSDs of polyacrylonitrile (PAN) and polyvinylidene difluoride (PVDF) HF membranes by evapoporometry (EP) characterization. The membranes were characterized before and after repeated cycles of filtration and backwashing in the absence of any foulants, and for a feed solution of bentonite and humic acid that caused fouling both on and within the membrane pores. Firstly, in the absence of any foulants, backwashing caused the appearance of larger pores, the effect of which was greater for the rubbery PVDF membrane than for the glassy PAN membrane. Secondly, backwashing was more effective in removing the fouling within the larger pores, but could not remove all the deposits within the smaller pores, which provides a mechanistic explanation for the progressive increase in the transmembrane pressure (TMP) with each backwashing cycle. Thirdly, the membranes that did not undergo the 10th backwashing at the end of 10 cycles of filtration and backwashing displayed a marked shift of the PSD towards smaller pores due to the deposition of foulants on and within the largest pores, whereas those that underwent the 10 complete backwashing cycles achieved nearly complete recovery of the larger pores accompanied by an irreversible increase in the diameter of the largest pores. Fourthly, a higher backwashing flux led to similar average pore diameters of the fouled and virgin membranes due to the increased effectiveness in restoring the smallest pores, but the corresponding higher filtration flux negated the benefits due to a greater fouling extent particularly for the larger pores. Finally, in order to achieve the desired permeation and rejection properties, possible enlargement of the pores needs to be taken into consideration when choosing a ultrafiltration (UF) membrane and when specifying the backwashing intensity and protocol. [ABSTRACT FROM AUTHOR]

The mechanical scouring of membranes induced by fluidized particles is a promising method for mitigating membrane fouling. In order to further improve this technique in terms of scouring efficiency and energy requirement, the goal of the current study is to understand the hydrodynamics of the fluidized granular activated carbon (GAC) particles, and correspondingly the relationship between the particle behavioral characteristics and the extent of membrane-fouling mitigation. In this study, the impact of GAC particle diameter ( d p ), superficial liquid velocity ( U l ) and axial position of the membrane ( h/H ) on particle behavior (namely, particle velocity, concentration and momentum) and membrane fouling was investigated. A high-speed video camera and the ImageJ software were used to characterize particle velocity and concentration, and membrane fouling was characterized by the rate of trans-membrane pressure (TMP) rise. Directly correlating particle behavior with membrane fouling trends shows that higher particle velocities, particle concentration and particle momentum generally all showed a negative correlation with dTMP/dt (i.e., improved fouling control). Of the three characteristics (namely, particle velocity, concentration and momentum), the negative correlation between dTMP/dt and particle momentum was the most pronounced, which suggests that momentum transfer between the GAC particles and the membrane represents a key mechanism effecting the scouring to diminish membrane fouling. Finally, the lack of a strong correlation between dTMP/dt and power input was observed, which indicates that a higher energy expenditure did not guarantee a more effective reduction in membrane fouling. [ABSTRACT FROM AUTHOR]

The Derjaguin-Landau-Verwey-Overbeek (DLVO) model, as well as the extended model (XDLVO), is popularly employed to quantify the interfacial interactions underlying membrane fouling. However, disagreements between membrane fouling extent and the interaction energies derived from DLVO or XDLVO models have been reported, which suggests gaps in the understanding. This study demonstrates that the DLVO approximation methods for predicting the interfacial foulant-membrane interaction are sensitive to the boundary condition assumptions (e.g., constant charge versus constant potential). In particular, while both the Poisson-Boltzmann (P–B) and linear superposition approximation (LSA) equations can quantify the electrostatics (EL) interaction energy component, the former assumes constant potential, while the latter additionally considers constant charge scenarios. The relative accuracy of these two equations were evaluated here. For dead-end filtration tests, flux decline trends, OCT analysis results and fouling model parameters were obtained. Regarding fouling of pristine PCTE membrane by latex particles of opposite charge signs, both the P–B and LSA equations contribute to predict relative fouling extents correctly. However, for the case of the BPEI-coated membrane, the P–B equation failed whereas the LSA equation gave good agreements. For cross-flow filtration tests in organic solvents, LSA also out-performed the P–B equation in providing more accurate predictions of membrane fouling. The results here highlight the shortcomings in the commonly used P–B equation and are expected to be potentially valuable in the development of better approximations for quantifying interfacial interaction energies. [Display omitted] • Both P–B and LSA equations predict fouling of pristine membrane by latex correctly. • Only LSA predict affinity between coated membrane layer and latex correctly. • Superiority of LSA also applies to fouling in organic solvents. • P–B equation is more limited due to underlying constant-potential assumption. • Boundary condition assumptions need to be considered in use of DLVO model. [ABSTRACT FROM AUTHOR]

This study mainly focused on enhancing oil-membrane electrostatic repulsion towards anti-oil-fouling via electrically conductive hydrophobic membrane in electricity-assisted membrane distillation (MD). Carbon nanotubes (CNTs) were coated on commercial membranes to fabricate membrane cathode. For concentrated hexadecane-in-saline water emulsions, the modified membrane exhibited significantly less flux decline, < 5% in dealing with extreme high-concentration oil emulsion (2000 ppm) at cell potential of 3.0 V. The anti-oil-fouling robustness was further confirmed over a continuous three-cycle operation with in-situ DI rinse. The anti-fouling mechanism was systematically discussed regarding the hydrophilicity of the membrane interface, charge repulsion between the oil and membrane cathode as well as slippery property at the liquid-gas-solid triple-phase interface. A modeled fouling rate constant was negatively associated with the calculated capacitive surface charge of the membrane cathode. Thermodynamics analysis suggested enhanced foulant-membrane electrostatic repulsion leads to a significant energy barrier which favored anti-fouling performance. Interestingly, the sliding dynamics of oil droplet along the interface of membrane cathode was found, tuned by the herein weak cell potential, which could also contribute to the fouling mitigation. Our results insight that the electrically conductive membrane cathode could modulate the foulant-membrane interaction which plays an important role in mitigating fouling/wetting occurred at the triple-phase interface. [Display omitted] • Conductive membrane cathode consisting of CNT on PVDF substrate is proposed. • Electrostatic repulsion between oil droplet and membrane is firstly studied in MD. • Negligible flux decline was found for hexadecane emulsion of 2000 ppm. • Sliding of oil droplet along the membrane cathode is found. • Foulant-membrane interacting dynamics is modulated. [ABSTRACT FROM AUTHOR]

The current study aimed to understand both external and internal membrane fouling by three proteins with different net charges, namely, negatively charged pepsin and bovine serum albumin (BSA), as well as positively charged lysozyme. Polycarbonate track-etched (PCTE) membranes were used. Per electrostatic attraction, the flux decline was the worst for lysozyme, which is attributed by the fouling model to the greatest pore blockage (α) and pore constriction (β), and by field-emission scanning electron microscope (FESEM) and optical coherence tomography (OCT) to the most extensive external fouling. Between pepsin and BSA, BSA gave worse flux decline despite its more negative net charge. The fouling model indicates that BSA gave greater pore blockage (α) and denser internal cake (R c / R m), while the quartz crystal microbalance with dissipation (QCM-D) indicates a rigid cake structure. Notably, despite monotonic flux decline with filtration, the OCT fouling voxel trends show significant fluctuations, which has not been reported before and thus signify the unique behavior of protein foulants in straight-through pores. Specifically, the trends below and above the −4.5 μm layer (i.e., 4.5 μm below the feed-membrane interface) are perfectly opposite, indicating the non-uniform protein deposits slipping downwards in the membrane pores as filtration progressed. The dynamic movements of the protein cakes unveiled here warrant more understanding in future studies. • Surprising fluctuations of OCT fouled voxels despite monotonically decreasing flux. • Dynamic downward motion of fouled protein deposit in the straight-through pores. • Delayed fluctuations by BSA tied to densest and most rigid cake. • Worst flux decline by lysozyme due to high pore blockage and pore constriction. [ABSTRACT FROM AUTHOR]

Membrane fouling, which is a key obstacle in implementing membrane technology, has been studied extensively for aqueous feeds. With increasing interests in organic solvent applications, a corresponding effort on understanding membrane fouling is warranted. This study employs molecular dynamics simulations to unveil the mechanisms underlying the different adsorption behaviors of dextran onto a polyacrylonitrile (PAN) membrane in three polar and protic solvents, namely, water, formamide and ethanol. The dextran-membrane separation distance is the lowest for water, followed by ethanol then formamide, which agrees with the worse flux decline for water relative to formamide observed experimentally. The greatest adsorption tendency in water is tied to the most attractive dextran-membrane interaction. On the other hand, the lower adsorption tendency in formamide and ethanol is linked to enhanced solvation of the dextran molecule and membrane, which deters dextran adsorption onto the membrane. Specifically, formamide, which leads to the least adsorption, exhibits the most attractive solvent-dextran and solvent-membrane interaction energies, the highest solvent-accessible surface area (SASA) for dextran, and also the highest density of solvent molecules in the solvation shell closest to the membrane. As for ethanol, it gives the highest density of solvent molecules in the solvation shell closest to a part of the dextran. The solvation of foulant and membrane by water deviates from that by other similarly polar and protic solvents, which has important implications in membrane fouling and highlights the need for enhancing the understanding of membrane fouling behaviors in organic solvents. [Display omitted] • Adsorption tendency of dextran on membrane is greatest in water and least in formamide. • Dextran-membrane interaction is most attractive in water. • Solvent-dextran and solvent-membrane interactions are most attractive in formamide. • Competitive solvent and dextran adsorption onto membrane governs fouling. • Greater solvation of foulant and membrane in ethanol and formamide reduces fouling. [ABSTRACT FROM AUTHOR]

In this study, a prototype microgranular adsorptive filtration (μGAF) system was constructed employing a 7-bore ceramic membrane as the primary membrane and either heated aluminum oxide particles (HAOPs) or powdered activated carbon (PAC) as the pre-deposited dynamic membrane (DM). The system was used to pre-treat membrane bioreactor (MBR) effluent from a full-scale MBR-reverse osmosis (RO) water reclamation plant. The downstream RO performance and membrane fouling potential of the treated effluent were then assessed. The results indicated that: (i) although PAC removed more overall EfOM than HAOPs did, HAOPs were more effective in removing biopolymers such as polysaccharides and proteins, (ii) HAOPs virtually eliminated fouling of the primary ceramic membrane, whereas considerable fouling (much of it irreversible) occurred when the feed was pretreated with PAC, (iii) HAOPs removed more than 90% of the phosphorus and fluoride from the feed, but PAC removed negligible amounts of these contaminants, and (iv) HAOPs-treated effluent resulted in only a 43% decline in RO permeate water flux over 5 d of continuous filtration, as opposed to 62% flux decline for untreated or PAC-treated effluent. This study thus demonstrates the effectiveness of the HAOPs-based μGAF process as a pre-treatment for improving downstream RO recovery. • A prototype μGAF system was used to treat an MBR effluent from a full-scale MBR-RO plant. • μGAF using HAOPs demonstrated an efficient removal of biopolymer, phosphorus, silica and fluoride. • HAOPs was more effective at mitigating downstream RO fouling compared to PAC. • HAOPs demonstrated the potential to enhance the overall performance of a water recovery process. [ABSTRACT FROM AUTHOR]

Membrane technology is increasingly becoming a promising alternative in the chemical and pharmaceutical industries, wherein organic solvents may form the continuous phase. Regarding the inevitable membrane fouling phenomenon, although the knowledge base is rich for feeds involving water, an analogous understanding for feeds involving organic solvents is limited. Accordingly, in this study, we systematically investigated the fouling behaviors of a model colloidal foulant (namely, silica) dispersed in water and five organic solvents (namely, methanol, ethanol, acetone, toluene and hexane) during ultrafiltration. The flux decline trends were clearly different. The XDLVO model and a fouling model were employed to extract mechanistic insights. Firstly, zeta potential alone was a poor indicator of the fouling extent. Secondly, solvents with high polarity (i.e., methanol, ethanol) had repulsive foulant-foulant and foulant-membrane interfacial interactions, which were beneficial in mitigating membrane fouling, leading to lesser flux decline and lower cake resistance. Thirdly, solvents with no or low polarity (i.e., n-hexane, toluene and acetone) had attractive interfacial interactions, which worsened membrane fouling. However, attractive foulant-foulant interaction was beneficial in augmenting shear-induced diffusion, which mitigated fouling. Fourthly, the fouling parameters extracted from the fouling model generally were lesser and greater respectively for the high-polarity and lower-polarity solvents, which agree with the interfacial interaction values and flux decline trends. The insights emanating from this study on membrane fouling in organic solvents are expected to be valuable in the design and operation of such emerging membrane-filtration systems. Image 1 • Fouling behaviors for same colloidal foulant and membrane differ among solvents. • Zeta potentials were poor indicators of membrane fouling extent. • Repulsive interactions tied to solvents with high polarity mitigated fouling. • Attractive interactions for solvents with no or low polarity worsened fouling. • Attractive foulant-foulant interactions mitigated fouling due to back-transport. [ABSTRACT FROM AUTHOR]

The internal fouling of membranes is typically presumed and inferred, but the direct characterization of representative samples is challenging. This study targeted to assess internal fouling using the Optical Coherence Tomography (OCT) and Evapoporometry (EP) techniques for real-time monitoring during filtration and off-line measurement of the pore size distributions (PSDs) of the fouled membranes, respectively. The results were validated by atomic force microscopy (AFM) measurements and also the three-mechanism fouling model. The foulant used was the well-studied bovine serum albumin (BSA), while the membranes were three commercially available polymeric microfiltration membranes with similar nominal pore sizes and porosities. Although the OCT affirmed the progressive worsening of internal fouling in real-time, the quantitative comparison of the extents of internal fouling among the three membranes was not possible. As for EP, it was able to quantitatively compare the pore size distributions and average pore diameters to ascertain the different extents of internal fouling. The phenomenological model available was effective in quantitatively comparing the extents of pore-constriction among the three membranes, while AFM tied the worst internal fouling to the most attractive BSA-membrane affinity. More in-depth understanding of internal fouling is warranted to not only recommend better membranes but also facilitate the advancement of models. • Characterized internal fouling of three polymeric membranes. • OCT revealed the evolution of internal fouling but comparison not possible. • EP revealed leftward shift of pore size distribution of fouled membrane. • Three-mechanism model quantified different fouling mechanisms. • AFM results revealed internal fouling tied to foulant-membrane interaction. [ABSTRACT FROM AUTHOR]

Chiral purity is becoming increasingly critical particularly for the pharmaceutical industry. Compared to the conventional techniques, membrane-based separation is promising as an easily scalable, efficient, continuous operation. Past studies have proven the efficacy of membranes for such separations, but the knowledge available on such processes remain limited. Since membrane fouling is well acknowledged to be one of the key impediments to more widespread implementation of membrane-filtration processes, this study was targeted at addressing the question of whether a relationship exists between the extents of membrane fouling and chiral purification. Another goal was to provide a more comprehensive understanding of the effect of various operating parameters (namely, chiral concentration, ratio of chiral selector to chiral concentrations, pH, feed medium and cross-flow velocity (CFV)) on the extent of chiral separation. The membrane was a non-chiral-selective polyethersulfone ultrafiltration membrane, the chiral selector was bovine serum albumin (BSA), while the chiral species was D,L-phenylalanine (Phe). Higher Phe concentrations decreased the separation factor and increased flux decline, while the highest separation factor was at an intermediate ratio of BSA to Phe concentration and an intermediate flux decline. Also, an intermediate pH of 9 gave the highest separation factor but extensive flux decline, and phosphate buffered saline (PBS) gave better separation and lesser flux decline than water as the feed medium. Regarding cross-flow velocity, the separation factor increased with CFV and decreased with flux decline. With respect to the relationship between the extents of chiral purification and flux decline, a poor correlation between the two existed, which suggests that improving membrane-based chiral separation requires means to improve the binding between the chiral selector and chiral species, rather than the conventional strategies to mitigate membrane fouling. • Poor correlation between separation factor and flux decline. • Binding between BSA and L-Phe played important role to enhance chiral separation. • Separation factor increased with cross-flow velocity and decreased with flux decline. • Separation factor and flux decline decreased as Phe concentration increased. • Varying CFV and pH significantly enhanced chiral separation. [ABSTRACT FROM AUTHOR]